WO2016187317A1 - Metabolic biomarkers for memory loss - Google Patents
Metabolic biomarkers for memory loss Download PDFInfo
- Publication number
- WO2016187317A1 WO2016187317A1 PCT/US2016/033100 US2016033100W WO2016187317A1 WO 2016187317 A1 WO2016187317 A1 WO 2016187317A1 US 2016033100 W US2016033100 W US 2016033100W WO 2016187317 A1 WO2016187317 A1 WO 2016187317A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- subject
- metabolite profile
- normal
- memory impairment
- metabolite
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/92—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving lipids, e.g. cholesterol, lipoproteins, or their receptors
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/68—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving proteins, peptides or amino acids
- G01N33/6893—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving proteins, peptides or amino acids related to diseases not provided for elsewhere
- G01N33/6896—Neurological disorders, e.g. Alzheimer's disease
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2405/00—Assays, e.g. immunoassays or enzyme assays, involving lipids
- G01N2405/04—Phospholipids, i.e. phosphoglycerides
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2800/00—Detection or diagnosis of diseases
- G01N2800/28—Neurological disorders
- G01N2800/2814—Dementia; Cognitive disorders
- G01N2800/2821—Alzheimer
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2800/00—Detection or diagnosis of diseases
- G01N2800/50—Determining the risk of developing a disease
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2800/00—Detection or diagnosis of diseases
- G01N2800/60—Complex ways of combining multiple protein biomarkers for diagnosis
Definitions
- the present invention relates to methods of determining if a subject has an increased risk of suffering from memory impairment.
- the methods comprise analyzing at least one plasma sample from the subject to determine a value of the subject's metabolite profile and comparing the value of the subject's metabolite profile with the value of a normal metabolite profile. A change in the value of the subject's metabolite profile, over or under normal values is indicative that the subject has an increased risk of suffering from memory impairment compared to a normal individual.
- AD Alzheimer's disease
- AD is a neurodegenerative disorder characterized by a progressive dementia that insidiously and inexorably robs older adults of their memory and other cognitive abilities.
- the prevalence of AD is expected to double every 20 years from 35.6 million individuals worldwide in 2010 to 115 million affected individuals by 2050. There is no cure and current therapies are unable to slow the progression of the disease.
- Biomarkers for early disease including cerebrospinal fluid (CSF) tau and ⁇ levels, structural and functional magnetic resonance imaging (MRI), and the recent use of brain positron emission tomography (PET) amyloid imaging, are of limited use as widespread screening tools since they provide diagnostic options that are either invasive (i.e., require lumbar puncture), time-consuming (i.e., several hours in a scanner for most comprehensive imaging protocols), expensive, or have been primarily validated in comparisons between normal subjects and those with the disease.
- CSF cerebrospinal fluid
- MRI structural and functional magnetic resonance imaging
- PET brain positron emission tomography
- the preclinical interval resulting in MCI or AD is known to be variable, multifactorial, and extends for at least 7-10 years prior to the emergence of clinical signs (prodromal or manifest disease).
- multimodal neurocognitive testing remains the most accurate, standardized, and widely used pre-mortem screening method to determine the presence or absence of clinical MCI or AD.
- the utility of strict cognitive assessment for preclinical stages of MCI or AD is limited, however, as this approach is not only time-consuming but is expected, by definition, to be normal in preclinical subjects.
- Neuropsychological testing is able to quantitatively delineate specific brain alterations from normal, such as memory, attention, language, visuoperceptual, and executive functions, which are typically not expected to be affected or closely scrutinized in the preclinical stages of disease. Thus, information obtained from multiple diagnostic studies will probably be most useful in defining the MCI/AD preclinical stages, including neuropsychological testing and some form(s) of biomarker(s). While CSF and neuroimaging have been used to define preclinical MCI/AD to date, their clinical utility as screening tools for asymptomatic individuals is not established.
- the present invention relates to methods of determining if a subject has an increased risk of suffering from memory impairment.
- the methods comprise analyzing at least one plasma sample from the subject to determine a value of the subject's metabolite profile and comparing the value of the subject's metabolite profile with the value of a normal metabolite profile. A change in the value of the subject's metabolite profile, over or under normal values is indicative that the subject has an increased risk of suffering from memory impairment compared to a normal individual.
- Figure 1 depicts a horizontal box and whisker plots of plasma 24 metabolite panel results for clinical groups in discovery and validation cohorts. Comparative range of plasma metabolite levels for the targeted discovery and validation studies are displayed, allowing appraisal of metabolite results in the cognitively intact Normal (bottom) versus Converterpre (top) groups.
- the box defines the interquartile range (IQR, mid 50% of data), with the vertical line within the box representing the median value.
- the whiskers define the upper and lower 25% limits of the data, while the dots represent outliers (>1.5 IQR lengths from the ends of the box).
- the NC group featured 53 subjects in the Discovery and 20 in the Validation cohorts, while the C pre group included 18 and 10 subjects, for Discovery and Validation cohorts, respectively.
- Figure 2 depicts the analytic representations to discriminate C pre from normal control (NC) subjects.
- Figure 2A shows that a receiver operating characteristic (ROC) curve area under the curve (AUC) for the discovery cohort was equal to 1.00 with 95% confidence interval (in parentheses) ranging from 1.00 to 1.00.
- Figure 2B shows that the ROC AUC for the validation cohort was 0.995, with 95% confidence interval (shaded area on plot) ranging from 0.981 to 1.00.
- ROC receiver operating characteristic
- FIG 3 depicts the calculated Plasma 24 metabolite Index (P24MI) results depicted as vertical boxplots based on the logistic regression model that distinguishes between C pre and NC groups.
- the boxes represent the interquartile range (IQR, 25%-75%), with bars extending from the boxes representing the upper and lower quartiles.
- Solid horizontal lines within the IQR represent the mean value, while the dashed horizontal lines represent the median value.
- the dots above/below each box represent outliers (>1.5 IQR lengths from the ends of the box).
- the higher Index Values (left vertical axis) are associated with an increased risk of phenoconversion to aMCI or AD, as seen the C pre subjects, with Confidence (right vertical axis) of predicting phenoconversion transitioning from 90% to 100% at an Index Value of 48.
- an Index Value > 49 represents a C pre individual and a risk of phenoconversion of 100% within the 5-year study range. Note the relatively low variability of the P24MI for both the NC group and the C pre group, with no overlap.
- Figure 4 depicts the performance of the Preclinical Alzheimer's Disease Index (PADI).
- Figure 4B depicts the survival analysis of baseline PADI on accumulative MCI/AD cases over 5 years adjusting for age, years of education, APOE4 carrier, global cognition, depression, and memory complaints.
- PADI Preclinical Alzheimer's Disease Index
- the present invention relates to methods of determining if a subject has an increased risk of suffering from memory impairment.
- the methods comprise analyzing at least one plasma sample from the subject to determine a value of the subject's metabolite profile and comparing the value of the subject's metabolite profile with the value of a normal metabolite profile. A change in the value of the subject's metabolite profile, over or under normal values is indicative that the subject has an increased risk of suffering from memory impairment compared to a normal individual.
- test subject indicates a mammal, in particular a human or non-human primate.
- the test subject may or may not be in need of an assessment of a predisposition to memory impairment.
- the test subject may have a condition or may have been exposed to injuries or conditions that are associated with memory impairment prior to applying the methods of the present invention.
- the test subject has not been identified as a subject that may have a condition or may have been exposed to injuries or conditions that are associated with memory impairment prior to applying the methods of the present invention.
- the phrase "memory impairment” means a measureable or perceivable decline or decrease in the subject's ability to recall past events.
- the term “past events” includes both recent (new) events (short-term memory) or events further back in time (long- term memory).
- the methods are used to assess an increased risk of short-term memory impairment.
- the methods are used to assess an increased risk in long-term memory impairment.
- the memory impairment can be age-related memory impairment.
- the memory impairment may also be disease-related memory impairment.
- Examples of disease- related memory impairment include but are not limited to Alzheimer's Disease, Parkinson's Disease, Multiple Sclerosis, Huntington's Disease, Pick's Disease, Progressive Supranuclear Palsy, Brain Tumor(s), Head Trauma, and Lyme Disease to name a few.
- the memory impairment is related to amnestic mild cognitive impairment (aMCI).
- aMCI amnestic mild cognitive impairment
- the memory impairment is related to Alzheimer's Disease.
- the root cause of the memory impairment is not necessarily critical to the methods of the present invention.
- the measureable or perceivable decline in the subject's ability to recall past events may be assessed clinically by a health care provider, such as a physician, physician's assistant, nurse, nurse practitioner, psychologist, psychiatrist, hospice provider, or any other provider that can assess a subject's memory.
- the measureable or perceivable decline in the subject's ability to recall past events may be assessed in a less formal, non-clinical manner, including but not limited to the subject himself or herself, acquaintances of the subject, employers of the subject and the like.
- the invention is not limited to a specific manner in which the subject's ability to recall past events is assessed. In fact, the methods of the invention can be implemented without the need to assess a subject's ability to recall past events. Of course, the methods of the present invention may also include assessing the subject's ability to assess past events one or more times, both before determining the subject's metabolite profile after determining the subject's metabolite profile at least one time.
- the methods may also comprise assessing the individual's Preclinical Alzheimer's Disease Index (PADI).
- PADI is an assessment of an individual's likelihood of developing memory impairment based on a novel, synthetic scoring valuation comprising a memory component and an attention component.
- the PADI of the individual is assessed prior to determining the subject's metabolite profile and the metabolite profile is then used to further stratify subjects that are considered at risk of suffering from memory impairment based on the PADI.
- the PADI of the individual is assessed after determining the subject's metabolite profile and the PADI is then used to further stratify subjects that are considered at risk of suffering from memory impairment based on the metabolite profile.
- the PADI profile and the metabolite profile are assessed in conjunction with one another such that the combined composite profile is determined prior to any initial patient population stratification.
- the PADI is calculated as a simple ratio between attention performance evaluations and memory performance evaluations. In one embodiment, the PADI is calculated as a weighted ratio between attention performance evaluations and memory performance evaluations. In specific embodiments, the attention component is weighted less than the memory component in calculating the PADI, i.e., the value of the attention is component less than 1:1, i.e., between about 0.001 and about 0.499. In more specific embodiments, the attention component is weighted between about 0.499 about 0.200, between about 0.490 and about 0.300 and between about 0.485 and about 0.400.
- the memory component is weighted more than the attention component in calculating the PADI, i.e., the value of the attention component is more than 1:1, i.e., between about 0.501 and about 0.999.
- the memory component is weighted between about 0.501 and 0.800, between about 0.510 and about 0.700 and between about 0.515 and about 0.600.
- the signs of relative weights in determining the ratios could be reversed for each component, e.g., the weight of the attention component could be between -0.001 and -0.500 and the weight of the memory component could be between about -.0500 and -0.999 such that the "higher" PADI score indicates that the subject has a higher risk of suffering from memory impairment.
- the subject's metabolite profile is combined with the subject's PADI to provide a composite score for assessing the risk that the subject will suffer from future memory impairment.
- select embodiments of the methods of the present invention comprise obtaining a metabolite profile, obtaining a PADI and combining each of the profiles to obtain a composite profile having a performance component and a biological component.
- the composite profile is calculated as a simple ratio between the PADI and the metabolite profile. In one embodiment, the composite profile is calculated as a weighted ratio between the PADI and the metabolite profile. In specific embodiments, the PADI component is weighted less than the metabolite component in calculating the composite profile, i.e., the value of the PADI component is less than 1:1, i.e., between about 0.001 and about 0.499. In more specific embodiments, the PADI component is weighted between about 0.499 about 0.200, between about 0.490 and about 0.300 and between about 0.485 and about 0.400.
- the PADI component is weighted more than the metabolite component in calculating the composite profile, i.e., the value of the PADI component is more than 1:1, i.e., between about 0.501 and about 0.999.
- the PADI component is weighted between about 0.501 and 0.800, between about 0.510 and about 0.700 and between about 0.515 and about 0.600.
- the metabolite component is weighted more than the PADI component in calculating the composite, i.e., the value of the PADI component is more than 1:1, i.e., between about 0.501 and about 0.999.
- the metabolite component is weighted between about 0.501 and 0.800, between about 0.510 and about 0.700 and between about 0.515 and about 0.600.
- the metabolite component is weighted less than the PADI component in calculating the composite profile, i.e., the value of the metabolite component is less than 1:1, i.e., between about 0.001 and about 0.499.
- the metabolite component is weighted between about 0.499 about 0.200, between about 0.490 and about 0.300 and between about 0.485 and about 0.400.
- the signs of relative weights in determining the composite could be reversed for each component, e.g., the weight of the PADI component could be between -0.001 and -0.500 and the weight of the metabolite component could be between about -.0500 and -0.999 such that the "higher" composite profile indicates that the subject has a higher risk of suffering from memory impairment.
- the decline or decrease in the ability to recall past events is relative to each individual's ability to recall past events prior to the diagnosed decrease or decline in the ability to recall past events. In another embodiment, the decline or decrease in the ability to recall past events is relative to a population's (general, specific or stratified) ability to recall past events prior to the diagnosed decrease or decline in the ability to recall past events.
- the term means "increased risk” is used to mean that the test subject has an increased chance of developing or acquiring memory impairment compared to a normal individual.
- the increased risk may be relative or absolute and may be expressed qualitatively or quantitatively.
- an increased risk may be expressed as simply determining the subject's metabolite profile and placing the patient in an "increased risk” category, based upon previous population studies.
- a numerical expression of the subject's increased risk may be determined based upon the metabolite profile.
- examples of expressions of an increased risk include but are not limited to, odds, probability, odds ratio, p-values, attributable risk, metabolite index score, relative frequency, positive predictive value, negative predictive value, and relative risk.
- the attributable risk can also be used to express an increased risk.
- the AR describes the proportion of individuals in a population exhibiting memory impairment due to a specific member of the metabolite risk profile. AR may also be important in quantifying the role of individual components (specific member) in disease etiology and in terms of the public health impact of the individual marker.
- the public health relevance of the AR measurement lies in estimating the proportion of cases of memory impairment in the population that could be prevented if the profile or individual component were absent.
- the increased risk of a patient can be determined from p-values that are derived from association studies. Specifically, associations with specific profiles can be performed using regression analysis by regressing the metabolite profile with memory impairment. In addition, the regression may or may not be corrected or adjusted for one or more factors.
- the factors for which the analyses may be adjusted include, but are not limited to age, sex, weight, ethnicity, geographic location, fasting state, state of pregnancy or post-pregnancy, menstrual cycle, general health of the subject, alcohol or drug consumption, caffeine or nicotine intake and circadian rhythms, and the subject's apolipoprotein E (ApoE) genotype to name a few.
- Increased risk can also be determined from p-values that are derived using logistic regression.
- Binomial (or binary) logistic regression is a form of regression which is used when the dependent is a dichotomy and the independents are of any type.
- Logistic regression can be used to predict a dependent variable on the basis of continuous and/or categorical independents and to determine the percent of variance in the dependent variable explained by the independents; to rank the relative importance of independents; to assess interaction effects; and to understand the impact of covariate control variables.
- Logistic regression applies maximum likelihood estimation after transforming the dependent into a "logit" variable (the natural log of the odds of the dependent occurring or not). In this way, logistic regression estimates the probability of a certain event occurring.
- SAS statistical analysis software
- Stata and SPSS general purpose package
- Discovery analysis software is a general purpose package (similar to Stata and SPSS) created by Jim Goodnight and N.C. State University colleagues. Ready-to-use procedures handle a wide range of statistical analyses, including but not limited to, analysis of variance, regression, categorical data analysis, multivariate analysis, survival analysis, psychometric analysis, cluster analysis, and nonparametric analysis.
- select embodiments of the present invention comprise the use of a computer comprising a processor and the computer is configured or programmed to generate one or more metabolite profiles and/or to determine statistical risk.
- the methods may also comprise displaying the one or more profiles and/or risk profiles on a screen that is communicatively connected to the computer.
- two different computers can be used: one computer configured or programmed to generate one or more metabolite profiles and a second computer configured or programmed to determine statistical risk. Each of these separate computers can be communicatively linked to its own display or to the same display.
- the phrase "metabolite profile” means the combination of a subject's metabolites found in the peripheral blood or portions thereof, such as but not limited to plasma or serum.
- the metabolite profile is a collection of measurements, such as but not limited to a quantity or concentration, for individual metabolites taken from a test sample of the subject.
- test samples or sources of components for the metabolite profile include, but are not limited to, biological fluids, which can be tested by the methods of the present invention described herein, and include but are not limited to whole blood, such as but not limited to peripheral blood, serum, plasma, cerebrospinal fluid, urine, amniotic fluid, lymph fluids, and various external secretions of the respiratory, intestinal and genitourinary tracts, tears, saliva, milk, white blood cells, myelomas and the like.
- Test samples to be assayed also include but are not limited to tissue specimens including normal and abnormal tissue.
- levels of the individual lipid components of the metabolite profile are assessed using mass spectrometry in conjunction with ultra- performance liquid chromatography (UPLC), high-performance liquid chromatography (HPLC), gas chromatography (GC), gas chromatography/mass spectroscopy (GC/MS), and UPLC to name a few.
- UPLC ultra- performance liquid chromatography
- HPLC high-performance liquid chromatography
- GC gas chromatography
- GC/MS gas chromatography/mass spectroscopy
- UPLC ultra- performance liquid chromatography
- Other methods of assessing levels of the individual components include biological methods, such as but not limited to ELISA assays.
- the assessment of the levels of the individual components of the metabolite profile can be expressed as absolute or relative values and may or may not be expressed in relation to another component, a standard an internal standard or another molecule of compound known to be in the sample. If the levels are assessed as relative to a standard or internal standard, the standard may be added to the test sample prior to, during or after sample processing.
- a sample is taken from the subject.
- the sample may or may not processed prior assaying levels of the components of the metabolite profile.
- whole blood may be taken from an individual and the blood sample may be processed, e.g., centrifuged, to isolate plasma or serum from the blood.
- the sample may or may not be stored, e.g., frozen, prior to processing or analysis.
- Individual components of the metabolite profile include but are not limited to proprionyl-L- carnitine (C3), lysophosphatidylcholine acyl C1S:2 (lysoPC a C18:2), phosphatidylcholine diacyl C36:6 (PC aa C36:6), phosphatidylcholine diacyl C38:0 (PC aa C38:0), phosphatidylcholine diacyl C38:6 (PC aa C38:6), phosphatidylcholine diacyl C40:l (PC aa C4Q:1), phosphatidylcholine acyi-alkyl C40:6 (PC ae C40:6), dodecenoy!-L-carnitine (C12:l), phosphatidylcholine diacyl C32:0 (PC aa C32:G), hexadecadieny!-L-carnitine (C12
- phosphatidylcholine diacyl C38:3 (PC aa C38:3), decenoyl-L-carnitine (C10:l), phosphatidylcholine acyi-alkyl C36:4 (PC ae C36:4), decadienyl-L-carnitine (C10:2), nonayl-L-carnitine (C9) and phosphatidylcholine acyi-alkyl C42:l (PC ae C42il).
- Metabolite C species e.g., C3. denote acyicarnitines (ACs).
- PC Phosphocholine
- metabolites display combined numbers of carbon atoms for their two acyl groups (snl and sn2 positions), e.g., C38, whereas the combined number of double bonds (unsaturation) is displayed after the colon, e.g., C38:6.
- Acyl group linkages to choline backbone for PCs feature ester (a) or ether (e) linkage, e.g., PC ae C36:4.
- the individual levels of each of the metabolites are lower than those compared to normal levels. In another embodiment, one, two, three, four, five, six, seven, eight, nine, ten, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 or 24 of the levels of each of the metabolites are lower than normal levels while others, if any, are higher than normal levels. In another embodiment, the individual levels of each of the metabolites are higher than those compared to normal levels. In another embodiment, one, two, three, four, five, six, seven, eight, nine, ten, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 or 24 of the levels of each of the metabolites are higher than normal levels while others, if any, are higher than normal levels.
- the levels of depletion or augmentation of the metabolites compared to normal levels can vary.
- the levels of any one or more of the metabolites is at least 1.05, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 lower than normal levels.
- the number of "times" the levels of a metabolite is lower or higher over normal can be a relative or absolute number of times.
- the levels of the metabolites may be normalized to a standard and these normalized levels can then be compared to one another to determine if a metabolite is lower or higher.
- the metabolite profile comprises at least two, three, four, five, six, seven, eight, nine, ten, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 or all 24 of the metabolites listed above. If two metabolites are used in generating the metabolite profile, any combination of the two listed above can be used. If three metabolites are used in generating the metabolite profile, any combination of three of the metabolites listed above can be used. If four metabolites are used in generating the metabolite profile, any combination of four of the metabolites listed above can be used. If five metabolites are used in generating the metabolite profile, any combination of five of the metabolites listed above can be used.
- any combination of six of the metabolites listed above can be used. If seven metabolites are used in generating the metabolite profile, any combination of seven of the metabolites listed above can be used. If eight metabolites are used in generating the metabolite profile, any combination of eight of the metabolites listed above can be used. If nine metabolites are used in generating the metabolite profile, any combination of nine of the metabolites listed above can be used. If ten metabolites are used in generating the metabolite profile, any combination of ten of the metabolites listed above can be used.
- any combination of 11 of the metabolites listed above can be used.
- 12 metabolites are used in generating the metabolite profile, any combination of 12 of the metabolites listed above can be used.
- 13 metabolites are used in generating the metabolite profile, any combination of 13 of the metabolites listed above can be used.
- 14 metabolites are used in generating the metabolite profile, any combination of 14 of the metabolites listed above can be used.
- 15 metabolites are used in generating the metabolite profile, any combination of 15 of the metabolites listed above can be used.
- 16 metabolites are used in generating the metabolite profile, any combination of 16 of the metabolites listed above can be used.
- any combination of 17 of the metabolites listed above can be used.
- 18 metabolites are used in generating the metabolite profile, any combination of 18 of the metabolites listed above can be used.
- 19 metabolites are used in generating the metabolite profile, any combination of 19 of the metabolites listed above can be used.
- 20 metabolites are used in generating the metabolite profile, any combination of 20 of the metabolites listed above can be used.
- 21 metabolites are used in generating the metabolite profile, any combination of 21 of the metabolites listed above can be used.
- 22 metabolites are used in generating the metabolite profile, any combination of 22 of the metabolites listed above can be used.
- 23 metabolites are used in generating the metabolite profile, any combination of 23 of the metabolites listed above can be used.
- all 24 metabolites can be used in generating the metabolite profile.
- Table 1 below lists an exemplary analysis of the metabolites.
- the normalized ratios depict relative abundance of metabolites in the C pre group (noted here as Con-pre) as compared to the NC group.
- Con-pre a ratio of one (1) indicates no change while values less than one indicate a negative change in the diagnostic group as compared to the NC.
- values greater than one indicate a positive change in the diagnostic group compared to NC.
- a negative log ratio e.g., PC ae C40:6, PC aa C40:l, PC aa C38:6, indicates that levels of these individual metabolites were lower in C pre compared to NC subjects.
- a positive log ratio e.g., C16:2, C12:l, C10:l, indicate that levels of these individual metabolites were higher in C pre compared to NC subjects.
- levels of individual components of any non-lipid component of the metabolite profile from test samples are well known to the skilled technician, and the invention is not limited by the means by which the components are assessed.
- levels of the individual components of the non-lipid portion of the profile are assessed using quantitative arrays, PC , Northern Blot analysis, Western Blot analysis, mass spectroscopy, high-performance liquid chromatography (HPLC, high performance gas chromatography (HPGC) and the like.
- Other methods of assessing levels of the individual components include biological methods, such as but not limited to ELISA assays.
- determining levels of, for example, a fragment of protein being analyzed may be sufficient to conclude or assess that an individual component of the metabolite profile, including the lipid and non-lipid portions of the metabolite profile, being analyzed is increased or decreased.
- the presence/absence/strength of a detectable signal may be sufficient to assess levels of metabolites.
- the subject's metabolite profile is compared to the profile that is deemed to be a normal metabolite profile.
- an individual or group of individuals may be first assessed for their ability to recall past events to establish that the individual or group of individuals has a normal or acceptable memory and/or cognitive abilities. Once established, the metabolite profile of the individual or group of individuals can then be determined to establish a "normal metabolite profile.”
- a normal metabolite profile can be ascertained from the same subject when the subject is deemed to possess normal cognitive abilities and no signs (clinical or otherwise) of memory impairment.
- a "normal" metabolite profile is assessed in the same subject from whom the sample is taken prior to the onset of measureable, perceivable or diagnosed memory impairment. That is, the term
- the present invention also includes methods of monitoring the progression of memory impairment in a subject, with the methods comprising determining the subject's metabolite profile at more than one time point.
- some embodiments of the methods of the present invention will comprise determining the subject's metabolite profile at two, three, four, five, six, seven, eight, nine, 10 or even more time points over a period of time, such as a year, two years, three, years, four years, five years, six years, seven years, eight years, nine years or even 10 years or longer.
- the methods of monitoring a subject's risk of having memory impairment would also include embodiments in which the subject's metabolite profile is assessed before and/or during and/or after treatment of memory impairment.
- the present invention also includes methods of monitoring the efficacy of treatment of memory impairment by assessing the subject's metabolite profile over the course of the treatment and after the treatment.
- the methods of monitoring the efficacy of treatment of memory impairment comprise determining the subject's metabolite profile at at least one, two, three, four, five, six, seven, eight, nine or 10 or more different time points prior to the receipt of treatment for memory impairment and subsequently determining the subject's metabolite profile at at least one, two, three, four, five, six, seven, eight, nine or 10 or more different time points after beginning of treatment for memory impairment, and determining the changes, if any, in the metabolite profile of the subject.
- the treatment may be any treatment designed to increase a subject's ability to recall past events, i.e., improve a subject's memory.
- a normal metabolite profile is assessed in a sample from a different subject or patient (from the subject being analyzed) and this different subject does not have or is not suspected of having memory impairment.
- the normal metabolite profile is assessed in a population of healthy individuals, the constituents of which display no memory impairment.
- the subject's metabolite profile can be compared to a normal metabolite profile generated from a single normal sample or a metabolite profile generated from more than one normal sample.
- measurements of the individual components e.g., concentration, ratio, log ratios etc.
- concentration, ratio, log ratios etc. of the normal metabolite profile can fall within a range of values, and values that do not fall within this "normal range” are said to be outside the normal range.
- These measurements may or may not be converted to a value, number, factor or score as compared to measurements in the "normal range.”
- a measurement for a specific metabolite that is below the normal range may be assigned a value or -1, -2, -3, etc., depending on the scoring system devised.
- the "metabolite profile value" can be a single value, number, factor or score given as an overall collective value to the individual molecular components of the profile, or to the categorical components, i.e., a phosphatidylcholine portion, an acylcarnitine portion, a biogenic amine portion and/or an amino acid portion.
- the metabolite value may simply be the overall score of each individual or categorical value.
- the metabolite profile value could be a useful single number or score, the actual value or magnitude of which could be an indication of the actual risk of memory impairment, e.g., the "more negative” the value, the greater the risk of memory impairment.
- the "metabolite profile value" can be a series of values, numbers, factors or scores given to the individual components of the overall profile.
- the "metabolite profile value" may be a combination of values, numbers, factors or scores given to individual components of the profile as well as values, numbers, factors or scores collectively given to a group of components, such as a phosphatidylcholine portion, an acylcarnitine portion, a biogenic amine portion and/or an amino acid portion.
- the metabolite profile value may comprise or consist of individual values, number, factors or scores for specific component as well as values, numbers, factors or scores for a group on components.
- individual values from the metabolites can be used to develop a single score, such as a "combined metabolite index,” which may utilize weighted scores from the individual component values reduced to a diagnostic number value.
- the combined metabolite index may also be generated using non-weighted scores from the individual component values.
- the threshold value would be or could be set by the combined metabolite index from one or more normal subjects.
- the value of the metabolite profile can be the collection of data from the individual measurements and need not be converted to a scoring system, such that the "metabolite profile value" is a collection of the individual measurements of the individual components of the profile.
- a subject is diagnosed of having an increased risk of suffering from memory impairment if the subject's 24, 23, 22, 21, 20, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, nine, eight, seven, six, five, four, three, two or even one of the metabolites herein are at abnormal levels.
- the attending health care provider may subsequently prescribe or institute a treatment program.
- the present invention also provides for methods of screening individuals as candidates for treatment of memory impairment.
- the attending healthcare worker may begin treatment, based on the subject's metabolite profile, before there are perceivable, noticeable or measurable signs of memory impairment in the individual.
- the invention provides methods of treating memory impairment in a subject in need thereof.
- the treatment methods include obtaining a subject's metabolite or composite profile as defined herein and prescribing a treatment regimen to the subject if the metabolite and/or composite profile indicate that the subject is at risk of suffering from memory impairment.
- the methods of treatment also include methods of monitoring the effectiveness of a treatment for memory impairment.
- a treatment regimen has been established, with or without the use of the methods of the present invention to assist in a diagnosis of memory impairment
- the methods of monitoring a subject's metabolite or composite profile over time can be used to assess the effectiveness of a memory impairment treatment.
- the subject's metabolite or composite profile can be assessed over time, including before, during and after treatments for memory impairment.
- the metabolite or composite profile can be monitored, with, for example, the normalization or decline in the values of the profile over time being indicative that the treatment may showing efficacy or may lead to improvement in the subjects memory and/or cognitive abilities.
- Suitable treatments for memory impairment that may be initiated in response to an indication that the subject is at risk of suffering from memory impairment include but are not limited to prescribing or administering fenofibtrate, cholinesterase inhibitors and memantine to the subject.
- cholinesterase inhibitors include but are not limited to donepezil, rivastigmine and galantamine.
- Another treatment includes but is not limited to SOUVENAIDTM, which is a mixture of omega-3 fatty acids, choline, uridine monophosphate and a mixture of antioxidants and B vitamins.
- Other treatments for memory impairment include consumption of coconut oil, ginger, green tea, fatty fish, cold-water fish such as but not limited to salmon, tuna, trout, mackerel and sardines, soy products, blueberries and other dark berries, green leafy vegetables and cruciferous vegetables such as but not limited to broccoli.
- R/OCAS Rochester/Orange County Aging Study
- Normative data for Z-score calculations were derived from the performance of the participants on each of the cognitive tests adjusted for age, education, sex, and visit.
- an(last), and Zvis(last) were defined as the age-gender-education-visit-adjusted robust Z-scores for the last available visit for each subject.
- the aMCI/AD group was defined as those participants whose adjusted Zmem was 1 IQR below the median at their last available visit, i.e., Zmem(last) ⁇ -1.35.
- the neurocognitive analyses were designed to demonstrate the general equivalence of the discovery and validation samples on clinical and cognitive measures. Separate Multivariate Analysis of Variance (MANOVA's) tests were used to examine discovery/validation group performance on the composite Z-scores and on self-report measures of memory complaints, memory related functional impairment, depressive symptoms, and a global measure of cognitive function.
- MANOVA's Multivariate Analysis of Variance
- biomarker sample (discovery, validation) was the independent variable and M MQ, IADL, GDS, and MMSE were the dependent variables.
- M MQ, IADL, GDS, and MMSE were the dependent variables.
- LC/MS-grade acetonitrile (ACN), Isopropanol (IPA), water and methanol were purchased from Fisher Scientific (New Jersey, USA). High purity formic acid (99%) was purchased from Thermo- Scientific (Rockford, IL).
- Debrisoquine, 4-Nitrobenzoic acid (4-NBA), Pro-Asn, Glycoursodeoxycholic acid, Malic acid were purchased from Sigma (St. Louis, MO, USA). All lipid standards including 14:0 LPA, 17:0 Ceramide, 12:0 LPC, 18:0 Lyso PI and PC(22:6/0:0) were procured from Avanti Polar Lipids Inc. (USA).
- the plasma samples were thawed on ice and vortexed.
- 25 ⁇ of plasma sample was mixed with 175 ⁇ of extraction buffer (25% acetonitrile in 40% methanol and 35% water) containing internal standards [10 ⁇ of debrisoquine (1 mg/mL), 50 ⁇ of 4, nitro- benzoic acid (1 mg/mL), 27.3 ⁇ of Ceramide (1 mg/mL) and 2.5 ⁇ of LPA (lysophosphatidic acid) (4 mg/mL) in 10 mL).
- the samples were incubated on ice for 10 minutes and centrifuged at 14,000 rpm at 4°C for 20 minutes. The supernatant was transferred to a fresh tube and dried under vacuum.
- the dried samples were reconstituted in 200 ⁇ of buffer containing 5% methanol, 1% acetonitrile and 94% water.
- the samples were centrifuged at 13,000 rpm for 20 minutes at 4°C to remove fine particulates. The supernatant was transferred to a glass vial for UPLC-ESI-Q-TOF-MS analysis.
- Each sample (2 ⁇ ) was injected onto a reverse-phase CSH C18 1.7 ⁇ 2.1x100 mm column using an Acquity H-class UPLC system (Waters Corporation, USA).
- the gradient mobile phase comprised of water containing 0.1% formic acid solution (Solvent A), 100% acetonitrile (Solvent B) and 10% acetonitrile in isopropanol (IPA) containing 0.1% formic acid and lOmM ammonium formate (Solvent C).
- Each sample was resolved for 13 minutes at a flow rate of 0.5 mL/min for 8 min and then 0.4 mL/min from 8 to 13 min.
- the UPLC gradient consisted of 98% A and 2% B for 0.5 min then a ramp of curve 6 to 60% B and 40% A from 0.5 min to 4.0 min, followed by a ramp of curve 6 to 98% B and 2% A from 4.0 to 8.0 min, then ramped to 5% B and 95% C from 9.0 min to 10.0 min at a flow rate of 0.4 ml/min, and finally to 98% A and 2% B from 11.0 min to 13 minutes.
- the column eluent was introduced directly into the mass spectrometer by electrospray ionization.
- Mass spectrometry was performed on a Quadrupole-Time of Flight (Q-TOF) instrument (Xevo G2 QTOF, Waters Corporation, USA) operating in either negative (ESI ) or positive (EST) electrospray ionization mode with a capillary voltage of 3200 V in positive mode and 2800 V in negative mode, and a sampling cone voltage of 30 V in both modes.
- the desolvation gas flow was set to 750 L h "1 and the temperature was set to 350°C while the source temperature was set at 120°C.
- pooled quality control (QC) samples (generated by taking an equal aliquot of all the samples included in the experiment) were run at the beginning of the sample queue for column conditioning and every ten injections thereafter to assess inconsistencies that are particularly evident in large batch acquisitions in terms of retention time drifts and variation in ion intensity over time.
- QC quality control
- Targeted metabolomic analysis of plasma sample was performed using the Biocrates Absolute-IDQ P180 (BIOCRATES, Life Science AG, Innsbruck, Austria). This validated targeted assay allows for simultaneous detection and quantification of metabolites in plasma samples (10 ⁇ ) in a high throughput manner. The methods have been described in detail.
- the plasma samples were processed as per the instructions by the manufacturer and analyzed on a triple quadrupole mass spectrometer (Xevo TQ.-S, Waters Corporation, USA) operating in the MRM mode.
- the flow injection analysis (FIA) tandem mass spectrometry (MS/MS) method was used to quantify a panel of 144 lipids simultaneously by multiple reaction monitoring.
- the other metabolites were resolved on the UPLC and quantified using scheduled MRMs.
- the kit facilitated absolute quantitation of 21 amino acids, hexose, carnitine, 39 acylcarnitines, 15 sphingomyelins, 90 phosphatidylcholines and 19 biogenic amines.
- Data pre-processing was performed using the MetlQ software (Biocrates) while the statistical analyses were performed using linear regression as well as the STAT pack module v3 (Biocrates). The concentration is expressed as nmol/L. Quality control samples were used to assess reproducibility of the assay.
- the mean of the coefficient of variation (CV) for the 180 metabolites was 0.08 and 95 % of the metabolites had a CV of ⁇ 0.15.
- Metabolite Statistical Analyses [0071] The m/z features of metabolites were normalized with log transformation that stabilized the variance followed with a quantile normalization to make the empirical distribution of intensities the same across samples. The metabolites were selected among all those known to be identifiable using a ROC regularized learning technique, based on the least absolute shrinkage and selection operator (LASSO) penalty as implemented with the R package 'gimnet', which uses cyclical coordinate descent in a pathwise fashion. The regularization path over a grid of values was obtained for the tuning parameter lambda through 10-fold cross-validation. The optimal value of the tuning parameter lambda, which was obtained by the cross-validation procedure, was then used to fit the model.
- LASSO least absolute shrinkage and selection operator
- the metabolomic data was used from the untargeted LASSO analysis to build separate classifier models that would distinguish the NC group from the C pre group.
- a receiver operating characteristic (ROC) analysis was employed to determine the a rea under the ROC curve (AUC) to assess the performance of the classifier models in differentiating the groups.
- ROC receiver operating characteristic
- AUC a rea under the ROC curve
- Alzheimer's disease is a cognitive disorder defined in numerous diagnostic schemes by a primary impairment in memory.
- the classification of preclinical Alzheimer's disease via PADI relies heavily on the assessment of memory, but it is recognized that the preclinical state may also involve other cognitive domains, thus the methods disclosed herein rely on assessing all primary cognitive domains (Table 4).
- the primary Memory outcome was based on the Rey Auditory Visual Learning Test (RAVLT). Learning was defined as the sum of the number of correct words recalled over the 5 learning trials; Retrieval was defined as the total number of correct words recalled from trial A7; and Retention was defined as the total number correct words recognized minus the number of false positives.
- RAVLT Rey Auditory Visual Learning Test
- Each of these three sub-scores was converted to an empirical Z-score by subtracting its sample mean and dividing by its sample SD.
- the three resulting positively correlated RAVLT-based Z-scores were then averaged to form the composite Memory score (Zmem). So defined, Zmem was approximately normal with mean 0, SD ⁇ 1, and some negative skewness attributable to the fact that healthy participants often score at or near the upper bounds, especially for Retention.
- the Attention (Zatt), Executive (Zexe), and Language (Zlan) scores were each defined as averages of empirical Z-score transforms of pairs of subscores, as follows. Zatt: (1) completion time (in seconds, truncated at 300s) for the Trail Making Test Part A and (2) the
- Wechsler Memory Scale-Ill Forward Digit Span Zexe: (1) completion time (in seconds, truncated at 300s) for the Trail Making Test Part B and (2) the Wechsler Memory Scale-Ill Backward Digit Span; Zlan: (1) total score (out of 60) for the Boston Naming Test and (2) Category Fluency (Animals Named in 1 minute).
- the Visuoperceptual score (Zvis) was simply the empirical Z-transform of the Total Score (out of 30) for the Hooper Visual Organization Test (HVOT), which itself was approximately normally distributed. See Table 2 for composite Z score components.
- Memory increased with years of education between 5-12 years, decreased between 12-14 years, increased again between 14-16 years, and then nearly leveled off for 16-23 years.
- Age was modeled linearly, as there was insufficient evidence to support nonlinearity via piecewise linear spline knots or a quadratic component, yet Memory scores decreased linearly with age.
- Visit was modeled via an indicator for baseline visit 0, given that there was insufficient evidence that subsequent visits differed from each other, yet there was evidence that subsequent visits had higher Memory scores compared with baseline visit 0.
- Gender was modeled via an indicator for males, who had lower memory scores than females. There was insufficient evidence to support any interactions.
- the choice to use robust measures of location (median) and dispersion (IQR) was made to reduce the influence that cognitively impaired participants might have on the mean and SD, and in recognition of the slight negative skewness of the residuals.
- the robustly standardized residuals were then viewed as age-gender-education-visit-adjusted robust Z-scores for each of the five cognitive domains.
- Zmem(last), Zatt(last), Zexe(last), Zlan(last), and Zvis(last) were defined as the age-gender-education-visit-adjusted robust Z-scores for the last available visit for each subject.
- the aMCI/AD group was defined as those participants whose adjusted Zmem was 1 IQR below the median at their last available visit, i.e., Zmem(last) ⁇ -1.35.
- NC participants were frequency matched to the aMCI/AD participants for each of the Discovery and Validation samples based on age, education, and sex. This was accomplished by stratifying the sample by four groups of age at the last visit (75-79, 80-85, 86-94, 95-100), three education groups (5-12, 13-18, 19-23), and two sexes (male, female), resulting in 24 age-education- sex strata.
- the PADI was constructed by combining the two domains (weighted -0.484 for attention and -0.516 for memory) such that higher scores indicate poorer cognitive performance.
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Hematology (AREA)
- Chemical & Material Sciences (AREA)
- Urology & Nephrology (AREA)
- Immunology (AREA)
- Analytical Chemistry (AREA)
- General Health & Medical Sciences (AREA)
- Microbiology (AREA)
- Biotechnology (AREA)
- Pathology (AREA)
- Cell Biology (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Biochemistry (AREA)
- Neurosurgery (AREA)
- Neurology (AREA)
- Proteomics, Peptides & Aminoacids (AREA)
- Biophysics (AREA)
- Endocrinology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The present invention relates to methods of determining if a subject has an increased risk of suffering from memory impairment. The methods comprise analyzing at least one plasma sample from the subject to determine a value of the subject's metabolite profile and comparing the value of the subject's metabolite profile with the value of a normal metabolite profile. A change in the value of the subject's metabolite profile, over normal values is indicative that the subject has an increased risk of suffering from memory impairment compared to a normal individual.
Description
METABOLIC BIOMARKERS FOR MEMORY LOSS Statement Regarding Federally Sponsored Research or Development
[0001] Part of the work performed during development of this invention utilized U.S. Government funds under National Instituted of Health Grant No. 01 AG030753 and Department of Defense Contract No. W81XWH-09-1-0107. The U.S. Government has certain rights in this invention.
Background of the Invention
Field of the Invention
[0002] The present invention relates to methods of determining if a subject has an increased risk of suffering from memory impairment. The methods comprise analyzing at least one plasma sample from the subject to determine a value of the subject's metabolite profile and comparing the value of the subject's metabolite profile with the value of a normal metabolite profile. A change in the value of the subject's metabolite profile, over or under normal values is indicative that the subject has an increased risk of suffering from memory impairment compared to a normal individual.
Background of the Invention
[0003] Alzheimer's disease (AD) is a neurodegenerative disorder characterized by a progressive dementia that insidiously and inexorably robs older adults of their memory and other cognitive abilities. The prevalence of AD is expected to double every 20 years from 35.6 million individuals worldwide in 2010 to 115 million affected individuals by 2050. There is no cure and current therapies are unable to slow the progression of the disease.
[0004] Early detection of the at-risk population (preclinical), or those in the initial symptomatic stages (prodromal, or mild cognitive impairment, MCI) of AD, may present opportunities for more successful therapeutic intervention, or even disease prevention by interdicting the
neuropathological cascade that is ultimately characterized by the deposition of extracellular β- amyloid (Αβ) and accumulation of intracellular fibrils of hyperphosphorylated microtubule associated protein tau (MAPT or tau) within the brain. Biomarkers for early disease, including cerebrospinal fluid (CSF) tau and Αβ levels, structural and functional magnetic resonance imaging (MRI), and the recent use of brain positron emission tomography (PET) amyloid imaging, are of limited use as widespread screening tools since they provide diagnostic options that are either invasive (i.e., require lumbar puncture), time-consuming (i.e., several hours in a scanner for most comprehensive imaging protocols), expensive, or have been primarily validated in comparisons between normal subjects and those with the disease. No current blood-based biomarkers can
detect incipient dementia with the required sensitivity and specificity during the preclinical stages. Continued interest in blood-based biomarkers remains because these specimens are obtained using minimally invasive, rapid, and relatively inexpensive methods. With recent technological advances in 'omics' technologies and systems biology analytic approaches, the comprehensive bioinformatic analyses of blood-based biomarkers may not only yield improved accuracy in predicting those at risk, but may also provide new insights into the underlying mechanisms and pathobiological networks involved in AD and possibly herald the development of new therapeutic strategies.
[0005] The preclinical interval resulting in MCI or AD is known to be variable, multifactorial, and extends for at least 7-10 years prior to the emergence of clinical signs (prodromal or manifest disease). In the absence of accurate and easily obtained biomarkers, multimodal neurocognitive testing remains the most accurate, standardized, and widely used pre-mortem screening method to determine the presence or absence of clinical MCI or AD. The utility of strict cognitive assessment for preclinical stages of MCI or AD is limited, however, as this approach is not only time-consuming but is expected, by definition, to be normal in preclinical subjects. Neuropsychological testing is able to quantitatively delineate specific brain alterations from normal, such as memory, attention, language, visuoperceptual, and executive functions, which are typically not expected to be affected or closely scrutinized in the preclinical stages of disease. Thus, information obtained from multiple diagnostic studies will probably be most useful in defining the MCI/AD preclinical stages, including neuropsychological testing and some form(s) of biomarker(s). While CSF and neuroimaging have been used to define preclinical MCI/AD to date, their clinical utility as screening tools for asymptomatic individuals is not established.
Summary of the Invention
[0006] The present invention relates to methods of determining if a subject has an increased risk of suffering from memory impairment. The methods comprise analyzing at least one plasma sample from the subject to determine a value of the subject's metabolite profile and comparing the value of the subject's metabolite profile with the value of a normal metabolite profile. A change in the value of the subject's metabolite profile, over or under normal values is indicative that the subject has an increased risk of suffering from memory impairment compared to a normal individual.
Brief Description of the Drawings
[0007] Figure 1 depicts a horizontal box and whisker plots of plasma 24 metabolite panel results for clinical groups in discovery and validation cohorts. Comparative range of plasma metabolite levels for the targeted discovery and validation studies are displayed, allowing appraisal of metabolite
results in the cognitively intact Normal (bottom) versus Converterpre (top) groups. The box defines the interquartile range (IQR, mid 50% of data), with the vertical line within the box representing the median value. The whiskers define the upper and lower 25% limits of the data, while the dots represent outliers (>1.5 IQR lengths from the ends of the box). The NC group featured 53 subjects in the Discovery and 20 in the Validation cohorts, while the Cpre group included 18 and 10 subjects, for Discovery and Validation cohorts, respectively. Individual analytes are listed on the left vertical axis, while normalized metabolite levels are shown on the horizontal axis. All the Cpre analyte results are reduced in comparison to NC levels, except for three acylcarnitine species (bottom of figure), C16:2, C12:l, and C10:l, which are elevated. Note the higher variability of the validation set compared to the discovery set, due to less than half of the number of subjects in the former compared to the latter.
[0008] Figure 2 depicts the analytic representations to discriminate Cpre from normal control (NC) subjects. Figure 2A shows that a receiver operating characteristic (ROC) curve area under the curve (AUC) for the discovery cohort was equal to 1.00 with 95% confidence interval (in parentheses) ranging from 1.00 to 1.00. Figure 2B shows that the ROC AUC for the validation cohort was 0.995, with 95% confidence interval (shaded area on plot) ranging from 0.981 to 1.00.
[0009] Figure 3 depicts the calculated Plasma 24 metabolite Index (P24MI) results depicted as vertical boxplots based on the logistic regression model that distinguishes between Cpre and NC groups. The boxes represent the interquartile range (IQR, 25%-75%), with bars extending from the boxes representing the upper and lower quartiles. Solid horizontal lines within the IQR represent the mean value, while the dashed horizontal lines represent the median value. The dots above/below each box represent outliers (>1.5 IQR lengths from the ends of the box). The higher Index Values (left vertical axis) are associated with an increased risk of phenoconversion to aMCI or AD, as seen the Cpre subjects, with Confidence (right vertical axis) of predicting phenoconversion transitioning from 90% to 100% at an Index Value of 48. Based on the calculated P24MI in the dataset, an Index Value > 49 represents a Cpre individual and a risk of phenoconversion of 100% within the 5-year study range. Note the relatively low variability of the P24MI for both the NC group and the Cpre group, with no overlap.
[0010] Figure 4 depicts the performance of the Preclinical Alzheimer's Disease Index (PADI). Figure 4A depicts the results of the ROC analysis of baseline PADI on phenoconversion to MCI/AD (AUC= 0.93); Figure 4B depicts the survival analysis of baseline PADI on accumulative MCI/AD cases over 5 years adjusting for age, years of education, APOE4 carrier, global cognition, depression, and memory complaints.
Detailed Description of the Invention
[0011] The present invention relates to methods of determining if a subject has an increased risk of suffering from memory impairment. The methods comprise analyzing at least one plasma sample from the subject to determine a value of the subject's metabolite profile and comparing the value of the subject's metabolite profile with the value of a normal metabolite profile. A change in the value of the subject's metabolite profile, over or under normal values is indicative that the subject has an increased risk of suffering from memory impairment compared to a normal individual.
[0012] As used herein, the term subject or "test subject" indicates a mammal, in particular a human or non-human primate. The test subject may or may not be in need of an assessment of a predisposition to memory impairment. For example, the test subject may have a condition or may have been exposed to injuries or conditions that are associated with memory impairment prior to applying the methods of the present invention. In another embodiment, the test subject has not been identified as a subject that may have a condition or may have been exposed to injuries or conditions that are associated with memory impairment prior to applying the methods of the present invention.
[0013] As used herein, the phrase "memory impairment" means a measureable or perceivable decline or decrease in the subject's ability to recall past events. As used herein, the term "past events" includes both recent (new) events (short-term memory) or events further back in time (long- term memory). In one embodiment, the methods are used to assess an increased risk of short-term memory impairment. In another embodiment, the methods are used to assess an increased risk in long-term memory impairment. The memory impairment can be age-related memory impairment. The memory impairment may also be disease-related memory impairment. Examples of disease- related memory impairment include but are not limited to Alzheimer's Disease, Parkinson's Disease, Multiple Sclerosis, Huntington's Disease, Pick's Disease, Progressive Supranuclear Palsy, Brain Tumor(s), Head Trauma, and Lyme Disease to name a few. In one embodiment, the memory impairment is related to amnestic mild cognitive impairment (aMCI). In another embodiment, the memory impairment is related to Alzheimer's Disease. The root cause of the memory impairment is not necessarily critical to the methods of the present invention. The measureable or perceivable decline in the subject's ability to recall past events may be assessed clinically by a health care provider, such as a physician, physician's assistant, nurse, nurse practitioner, psychologist, psychiatrist, hospice provider, or any other provider that can assess a subject's memory. The measureable or perceivable decline in the subject's ability to recall past events may be assessed in a less formal, non-clinical manner, including but not limited to the subject himself or herself,
acquaintances of the subject, employers of the subject and the like. The invention is not limited to a specific manner in which the subject's ability to recall past events is assessed. In fact, the methods of the invention can be implemented without the need to assess a subject's ability to recall past events. Of course, the methods of the present invention may also include assessing the subject's ability to assess past events one or more times, both before determining the subject's metabolite profile after determining the subject's metabolite profile at least one time.
[0014] In one embodiment, the methods may also comprise assessing the individual's Preclinical Alzheimer's Disease Index (PADI). As used herein, PADI is an assessment of an individual's likelihood of developing memory impairment based on a novel, synthetic scoring valuation comprising a memory component and an attention component. In one embodiment, the PADI of the individual is assessed prior to determining the subject's metabolite profile and the metabolite profile is then used to further stratify subjects that are considered at risk of suffering from memory impairment based on the PADI. In another embodiment, the PADI of the individual is assessed after determining the subject's metabolite profile and the PADI is then used to further stratify subjects that are considered at risk of suffering from memory impairment based on the metabolite profile. In yet another embodiment, the PADI profile and the metabolite profile are assessed in conjunction with one another such that the combined composite profile is determined prior to any initial patient population stratification.
[0015] In one embodiment, the PADI is calculated as a simple ratio between attention performance evaluations and memory performance evaluations. In one embodiment, the PADI is calculated as a weighted ratio between attention performance evaluations and memory performance evaluations. In specific embodiments, the attention component is weighted less than the memory component in calculating the PADI, i.e., the value of the attention is component less than 1:1, i.e., between about 0.001 and about 0.499. In more specific embodiments, the attention component is weighted between about 0.499 about 0.200, between about 0.490 and about 0.300 and between about 0.485 and about 0.400. Conversely, in some embodiments, the memory component is weighted more than the attention component in calculating the PADI, i.e., the value of the attention component is more than 1:1, i.e., between about 0.501 and about 0.999. For example, in more specific embodiments, the memory component is weighted between about 0.501 and 0.800, between about 0.510 and about 0.700 and between about 0.515 and about 0.600. Of course, the signs of relative weights in determining the ratios could be reversed for each component, e.g., the weight of the attention component could be between -0.001 and -0.500 and the weight of the memory
component could be between about -.0500 and -0.999 such that the "higher" PADI score indicates that the subject has a higher risk of suffering from memory impairment.
[0016] In specific embodiments, the subject's metabolite profile is combined with the subject's PADI to provide a composite score for assessing the risk that the subject will suffer from future memory impairment. Accordingly, select embodiments of the methods of the present invention comprise obtaining a metabolite profile, obtaining a PADI and combining each of the profiles to obtain a composite profile having a performance component and a biological component.
[0017] In one embodiment, the composite profile is calculated as a simple ratio between the PADI and the metabolite profile. In one embodiment, the composite profile is calculated as a weighted ratio between the PADI and the metabolite profile. In specific embodiments, the PADI component is weighted less than the metabolite component in calculating the composite profile, i.e., the value of the PADI component is less than 1:1, i.e., between about 0.001 and about 0.499. In more specific embodiments, the PADI component is weighted between about 0.499 about 0.200, between about 0.490 and about 0.300 and between about 0.485 and about 0.400. In additional embodiments, the PADI component is weighted more than the metabolite component in calculating the composite profile, i.e., the value of the PADI component is more than 1:1, i.e., between about 0.501 and about 0.999. For example, in more specific embodiments, the PADI component is weighted between about 0.501 and 0.800, between about 0.510 and about 0.700 and between about 0.515 and about 0.600. Conversely, in some embodiments, the metabolite component is weighted more than the PADI component in calculating the composite, i.e., the value of the PADI component is more than 1:1, i.e., between about 0.501 and about 0.999. For example, in more specific embodiments, the metabolite component is weighted between about 0.501 and 0.800, between about 0.510 and about 0.700 and between about 0.515 and about 0.600. In additional embodiments, the metabolite component is weighted less than the PADI component in calculating the composite profile, i.e., the value of the metabolite component is less than 1:1, i.e., between about 0.001 and about 0.499. In more specific embodiments, the metabolite component is weighted between about 0.499 about 0.200, between about 0.490 and about 0.300 and between about 0.485 and about 0.400. Of course, the signs of relative weights in determining the composite could be reversed for each component, e.g., the weight of the PADI component could be between -0.001 and -0.500 and the weight of the metabolite component could be between about -.0500 and -0.999 such that the "higher" composite profile indicates that the subject has a higher risk of suffering from memory impairment.
[0018] In one embodiment, the decline or decrease in the ability to recall past events is relative to each individual's ability to recall past events prior to the diagnosed decrease or decline in the ability
to recall past events. In another embodiment, the decline or decrease in the ability to recall past events is relative to a population's (general, specific or stratified) ability to recall past events prior to the diagnosed decrease or decline in the ability to recall past events.
[0019] As used herein, the term means "increased risk" is used to mean that the test subject has an increased chance of developing or acquiring memory impairment compared to a normal individual. The increased risk may be relative or absolute and may be expressed qualitatively or quantitatively. For example, an increased risk may be expressed as simply determining the subject's metabolite profile and placing the patient in an "increased risk" category, based upon previous population studies. Alternatively, a numerical expression of the subject's increased risk may be determined based upon the metabolite profile. As used herein, examples of expressions of an increased risk include but are not limited to, odds, probability, odds ratio, p-values, attributable risk, metabolite index score, relative frequency, positive predictive value, negative predictive value, and relative risk.
[0020] For example, the correlation between a subject's metabolite profile and the likelihood of suffering from memory impairment may be measured by an odds ratio (OR) and by the relative risk (RR). If P(R+) is the probability of developing memory impairment for individuals with the risk profile (R) and P(R") is the probability of developing memory impairment for individuals without the risk profile, then the relative risk is the ratio of the two probabilities: RR=P(R+)/P(R").
[0021] In case-control studies, however, direct measures of the relative risk often cannot be obtained because of sampling design. The odds ratio allows for an approximation of the relative risk for low-incidence diseases and can be calculated:
where F+ is the frequency of a risk profile in cases studies and F" is the frequency of risk profile in controls. F+ and F" can be calculated using the metabolite profile frequencies of the study.
[0022] The attributable risk (AR) can also be used to express an increased risk. The AR describes the proportion of individuals in a population exhibiting memory impairment due to a specific member of the metabolite risk profile. AR may also be important in quantifying the role of individual components (specific member) in disease etiology and in terms of the public health impact of the individual marker. The public health relevance of the AR measurement lies in estimating the proportion of cases of memory impairment in the population that could be prevented if the profile or individual component were absent. AR may be determined as follows: AR=PE(RR-1)/(PE(RR-1)+1), where AR is the risk attributable to a profile or individual component of the profile, and PE is the frequency of exposure to a profile or individual component of the profile within the population at large. RR is the relative risk, which can be approximated with the odds ratio when the profile or
individual component of the profile under study has a relatively low incidence in the general population.
[0023] In one embodiment, the increased risk of a patient can be determined from p-values that are derived from association studies. Specifically, associations with specific profiles can be performed using regression analysis by regressing the metabolite profile with memory impairment. In addition, the regression may or may not be corrected or adjusted for one or more factors. The factors for which the analyses may be adjusted include, but are not limited to age, sex, weight, ethnicity, geographic location, fasting state, state of pregnancy or post-pregnancy, menstrual cycle, general health of the subject, alcohol or drug consumption, caffeine or nicotine intake and circadian rhythms, and the subject's apolipoprotein E (ApoE) genotype to name a few.
[0024] Increased risk can also be determined from p-values that are derived using logistic regression. Binomial (or binary) logistic regression is a form of regression which is used when the dependent is a dichotomy and the independents are of any type. Logistic regression can be used to predict a dependent variable on the basis of continuous and/or categorical independents and to determine the percent of variance in the dependent variable explained by the independents; to rank the relative importance of independents; to assess interaction effects; and to understand the impact of covariate control variables. Logistic regression applies maximum likelihood estimation after transforming the dependent into a "logit" variable (the natural log of the odds of the dependent occurring or not). In this way, logistic regression estimates the probability of a certain event occurring. These analyses are conducted with the program SAS.
[0025] SAS ("statistical analysis software") is a general purpose package (similar to Stata and SPSS) created by Jim Goodnight and N.C. State University colleagues. Ready-to-use procedures handle a wide range of statistical analyses, including but not limited to, analysis of variance, regression, categorical data analysis, multivariate analysis, survival analysis, psychometric analysis, cluster analysis, and nonparametric analysis.
[0026] Accordingly, select embodiments of the present invention comprise the use of a computer comprising a processor and the computer is configured or programmed to generate one or more metabolite profiles and/or to determine statistical risk. The methods may also comprise displaying the one or more profiles and/or risk profiles on a screen that is communicatively connected to the computer. In another embodiment, two different computers can be used: one computer configured or programmed to generate one or more metabolite profiles and a second computer configured or
programmed to determine statistical risk. Each of these separate computers can be communicatively linked to its own display or to the same display.
[0027] As used herein, the phrase "metabolite profile" means the combination of a subject's metabolites found in the peripheral blood or portions thereof, such as but not limited to plasma or serum. The metabolite profile is a collection of measurements, such as but not limited to a quantity or concentration, for individual metabolites taken from a test sample of the subject. Examples of test samples or sources of components for the metabolite profile include, but are not limited to, biological fluids, which can be tested by the methods of the present invention described herein, and include but are not limited to whole blood, such as but not limited to peripheral blood, serum, plasma, cerebrospinal fluid, urine, amniotic fluid, lymph fluids, and various external secretions of the respiratory, intestinal and genitourinary tracts, tears, saliva, milk, white blood cells, myelomas and the like. Test samples to be assayed also include but are not limited to tissue specimens including normal and abnormal tissue.
[0028] Techniques to assay levels of individual components of the metabolite profile from test samples are well known to the skilled technician, and the invention is not limited by the means by which the components are assessed. In one embodiment, levels of the individual lipid components of the metabolite profile are assessed using mass spectrometry in conjunction with ultra- performance liquid chromatography (UPLC), high-performance liquid chromatography (HPLC), gas chromatography (GC), gas chromatography/mass spectroscopy (GC/MS), and UPLC to name a few. Other methods of assessing levels of the individual components include biological methods, such as but not limited to ELISA assays.
[0029] The assessment of the levels of the individual components of the metabolite profile can be expressed as absolute or relative values and may or may not be expressed in relation to another component, a standard an internal standard or another molecule of compound known to be in the sample. If the levels are assessed as relative to a standard or internal standard, the standard may be added to the test sample prior to, during or after sample processing.
[0030] To assess levels of the individual components of the metabolite profile, a sample is taken from the subject. The sample may or may not processed prior assaying levels of the components of the metabolite profile. For example, whole blood may be taken from an individual and the blood sample may be processed, e.g., centrifuged, to isolate plasma or serum from the blood. The sample may or may not be stored, e.g., frozen, prior to processing or analysis.
[0031] Individual components of the metabolite profile include but are not limited to proprionyl-L- carnitine (C3), lysophosphatidylcholine acyl C1S:2 (lysoPC a C18:2), phosphatidylcholine diacyl C36:6 (PC aa C36:6), phosphatidylcholine diacyl C38:0 (PC aa C38:0), phosphatidylcholine diacyl C38:6 (PC aa C38:6), phosphatidylcholine diacyl C40:l (PC aa C4Q:1), phosphatidylcholine acyi-alkyl C40:6 (PC ae C40:6), dodecenoy!-L-carnitine (C12:l), phosphatidylcholine diacyl C32:0 (PC aa C32:G), hexadecadieny!-L-carnitine (C16:2), phosphatidylcholine diacyl C40:5 (PC aa C4Q:5), hydroxyva!eryi-L- carnitine/methylma!onyl-L-carnitine (C5-OH/C3-DC-M), phosphatidylcholine acyl-a!kyi C34:0 (PC ae C34:0), hydroxyoctadecenoyl-L-carnitine (C18:1-QH), phosphatidylcholine diacyl C34:4 (PC aa C34:4), Asparagine (Asn), Asymmetric dimethylarginine (ADMA), valeryl-L-carnitine (C5),
phosphatidylcholine diacyl C38:3 (PC aa C38:3), decenoyl-L-carnitine (C10:l), phosphatidylcholine acyi-alkyl C36:4 (PC ae C36:4), decadienyl-L-carnitine (C10:2), nonayl-L-carnitine (C9) and phosphatidylcholine acyi-alkyl C42:l (PC ae C42il). Metabolite C species, e.g., C3. denote acyicarnitines (ACs). Phosphocholine (PC) metabolites display combined numbers of carbon atoms for their two acyl groups (snl and sn2 positions), e.g., C38, whereas the combined number of double bonds (unsaturation) is displayed after the colon, e.g., C38:6. Acyl group linkages to choline backbone for PCs feature ester (a) or ether (e) linkage, e.g., PC ae C36:4.
[0032] In one embodiment, the individual levels of each of the metabolites are lower than those compared to normal levels. In another embodiment, one, two, three, four, five, six, seven, eight, nine, ten, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 or 24 of the levels of each of the metabolites are lower than normal levels while others, if any, are higher than normal levels. In another embodiment, the individual levels of each of the metabolites are higher than those compared to normal levels. In another embodiment, one, two, three, four, five, six, seven, eight, nine, ten, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 or 24 of the levels of each of the metabolites are higher than normal levels while others, if any, are higher than normal levels.
[0033] The levels of depletion or augmentation of the metabolites compared to normal levels can vary. In one embodiment, the levels of any one or more of the metabolites is at least 1.05, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 lower than normal levels. For the purposes of the present invention, the number of "times" the levels of a metabolite is lower or higher over normal can be a relative or absolute number of times. In the alternative, the levels of the metabolites may be normalized to a standard and these normalized levels can then be compared to one another to determine if a metabolite is lower or higher.
[0034] For the purposes of the present invention the metabolite profile comprises at least two, three, four, five, six, seven, eight, nine, ten, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 or all 24
of the metabolites listed above. If two metabolites are used in generating the metabolite profile, any combination of the two listed above can be used. If three metabolites are used in generating the metabolite profile, any combination of three of the metabolites listed above can be used. If four metabolites are used in generating the metabolite profile, any combination of four of the metabolites listed above can be used. If five metabolites are used in generating the metabolite profile, any combination of five of the metabolites listed above can be used. If six metabolites are used in generating the metabolite profile, any combination of six of the metabolites listed above can be used. If seven metabolites are used in generating the metabolite profile, any combination of seven of the metabolites listed above can be used. If eight metabolites are used in generating the metabolite profile, any combination of eight of the metabolites listed above can be used. If nine metabolites are used in generating the metabolite profile, any combination of nine of the metabolites listed above can be used. If ten metabolites are used in generating the metabolite profile, any combination of ten of the metabolites listed above can be used. If 11 metabolites are used in generating the metabolite profile, any combination of 11 of the metabolites listed above can be used. If 12 metabolites are used in generating the metabolite profile, any combination of 12 of the metabolites listed above can be used. If 13 metabolites are used in generating the metabolite profile, any combination of 13 of the metabolites listed above can be used. If 14 metabolites are used in generating the metabolite profile, any combination of 14 of the metabolites listed above can be used. If 15 metabolites are used in generating the metabolite profile, any combination of 15 of the metabolites listed above can be used. If 16 metabolites are used in generating the metabolite profile, any combination of 16 of the metabolites listed above can be used. If 17 metabolites are used in generating the metabolite profile, any combination of 17 of the metabolites listed above can be used. If 18 metabolites are used in generating the metabolite profile, any combination of 18 of the metabolites listed above can be used. If 19 metabolites are used in generating the metabolite profile, any combination of 19 of the metabolites listed above can be used. If 20 metabolites are used in generating the metabolite profile, any combination of 20 of the metabolites listed above can be used. If 21 metabolites are used in generating the metabolite profile, any combination of 21 of the metabolites listed above can be used. If 22 metabolites are used in generating the metabolite profile, any combination of 22 of the metabolites listed above can be used. If 23 metabolites are used in generating the metabolite profile, any combination of 23 of the metabolites listed above can be used. Of course, all 24 metabolites can be used in generating the metabolite profile.
[0035] Table 1 below lists an exemplary analysis of the metabolites. The normalized ratios depict relative abundance of metabolites in the Cpre group (noted here as Con-pre) as compared to the NC group. Herein, a ratio of one (1) indicates no change while values less than one indicate a negative
change in the diagnostic group as compared to the NC. Herein, values greater than one indicate a positive change in the diagnostic group compared to NC. A negative log ratio, e.g., PC ae C40:6, PC aa C40:l, PC aa C38:6, indicates that levels of these individual metabolites were lower in Cpre compared to NC subjects. A positive log ratio, e.g., C16:2, C12:l, C10:l, indicate that levels of these individual metabolites were higher in Cpre compared to NC subjects.
[0036] Techniques to assay levels of individual components of any non-lipid component of the metabolite profile from test samples are well known to the skilled technician, and the invention is not limited by the means by which the components are assessed. In one embodiment, levels of the individual components of the non-lipid portion of the profile are assessed using quantitative arrays,
PC , Northern Blot analysis, Western Blot analysis, mass spectroscopy, high-performance liquid chromatography (HPLC, high performance gas chromatography (HPGC) and the like. Other methods of assessing levels of the individual components include biological methods, such as but not limited to ELISA assays. To determine levels of metabolites, it is not necessary that an entire metabolite, e.g., a full length protein or an entire RNA transcript, be present or fully sequenced. In other words, determining levels of, for example, a fragment of protein being analyzed may be sufficient to conclude or assess that an individual component of the metabolite profile, including the lipid and non-lipid portions of the metabolite profile, being analyzed is increased or decreased. Similarly, if, for example, arrays or blots are used to determine metabolite levels, the presence/absence/strength of a detectable signal may be sufficient to assess levels of metabolites.
[0037] The subject's metabolite profile is compared to the profile that is deemed to be a normal metabolite profile. To establish the metabolite profile of a normal individual, an individual or group of individuals may be first assessed for their ability to recall past events to establish that the individual or group of individuals has a normal or acceptable memory and/or cognitive abilities. Once established, the metabolite profile of the individual or group of individuals can then be determined to establish a "normal metabolite profile." In one embodiment, a normal metabolite profile can be ascertained from the same subject when the subject is deemed to possess normal cognitive abilities and no signs (clinical or otherwise) of memory impairment. In one embodiment, a "normal" metabolite profile is assessed in the same subject from whom the sample is taken prior to the onset of measureable, perceivable or diagnosed memory impairment. That is, the term
"normal" with respect to a metabolite profile can be used to mean the subject's baseline metabolite profile prior to the onset of memory impairment. The metabolite profile can then be reassessed periodically and compared to the subject's baseline metabolite profile. Thus, the present invention also includes methods of monitoring the progression of memory impairment in a subject, with the methods comprising determining the subject's metabolite profile at more than one time point. For example, some embodiments of the methods of the present invention will comprise determining the subject's metabolite profile at two, three, four, five, six, seven, eight, nine, 10 or even more time points over a period of time, such as a year, two years, three, years, four years, five years, six years, seven years, eight years, nine years or even 10 years or longer. The methods of monitoring a subject's risk of having memory impairment would also include embodiments in which the subject's metabolite profile is assessed before and/or during and/or after treatment of memory impairment. In other words, the present invention also includes methods of monitoring the efficacy of treatment of memory impairment by assessing the subject's metabolite profile over the course of the treatment and after the treatment. In specific embodiments, the methods of monitoring the
efficacy of treatment of memory impairment comprise determining the subject's metabolite profile at at least one, two, three, four, five, six, seven, eight, nine or 10 or more different time points prior to the receipt of treatment for memory impairment and subsequently determining the subject's metabolite profile at at least one, two, three, four, five, six, seven, eight, nine or 10 or more different time points after beginning of treatment for memory impairment, and determining the changes, if any, in the metabolite profile of the subject. The treatment may be any treatment designed to increase a subject's ability to recall past events, i.e., improve a subject's memory.
[0038] In another embodiment, a normal metabolite profile is assessed in a sample from a different subject or patient (from the subject being analyzed) and this different subject does not have or is not suspected of having memory impairment. In still another embodiment, the normal metabolite profile is assessed in a population of healthy individuals, the constituents of which display no memory impairment. Thus, the subject's metabolite profile can be compared to a normal metabolite profile generated from a single normal sample or a metabolite profile generated from more than one normal sample.
[0039] Of course, measurements of the individual components, e.g., concentration, ratio, log ratios etc., of the normal metabolite profile can fall within a range of values, and values that do not fall within this "normal range" are said to be outside the normal range. These measurements may or may not be converted to a value, number, factor or score as compared to measurements in the "normal range." For example, a measurement for a specific metabolite that is below the normal range, may be assigned a value or -1, -2, -3, etc., depending on the scoring system devised.
[0040] In one embodiment, the "metabolite profile value" can be a single value, number, factor or score given as an overall collective value to the individual molecular components of the profile, or to the categorical components, i.e., a phosphatidylcholine portion, an acylcarnitine portion, a biogenic amine portion and/or an amino acid portion. For example, if each component is assigned a value, such as above, the metabolite value may simply be the overall score of each individual or categorical value. For example, if 13 of the components of the metabolite profile are phosphatidylcholine, and seven of those components are assigned values of "-2" and six are assigned values of "-1," the phosphatidylcholine portion of the metabolite profile in this example would be -20, with a normal value being, for example, "0." In this manner, the metabolite profile value could be a useful single number or score, the actual value or magnitude of which could be an indication of the actual risk of memory impairment, e.g., the "more negative" the value, the greater the risk of memory impairment.
[0041] In another embodiment the "metabolite profile value" can be a series of values, numbers, factors or scores given to the individual components of the overall profile. In another embodiment, the "metabolite profile value" may be a combination of values, numbers, factors or scores given to individual components of the profile as well as values, numbers, factors or scores collectively given to a group of components, such as a phosphatidylcholine portion, an acylcarnitine portion, a biogenic amine portion and/or an amino acid portion. In another example, the metabolite profile value may comprise or consist of individual values, number, factors or scores for specific component as well as values, numbers, factors or scores for a group on components.
[0042] In another embodiment individual values from the metabolites can be used to develop a single score, such as a "combined metabolite index," which may utilize weighted scores from the individual component values reduced to a diagnostic number value. The combined metabolite index may also be generated using non-weighted scores from the individual component values. When the "combined metabolite index" exceeds a specific threshold level, determined by a range of values developed similarly from control subjects, the individual has a high risk, or higher than normal risk, of developing memory impairment, whereas maintaining a normal range value of the "combined metabolite index" would indicate a low or minimal risk of developing memory impairment. In this embodiment, the threshold value would be or could be set by the combined metabolite index from one or more normal subjects.
[0043] In another embodiment, the value of the metabolite profile can be the collection of data from the individual measurements and need not be converted to a scoring system, such that the "metabolite profile value" is a collection of the individual measurements of the individual components of the profile.
[0044] In specific embodiments, a subject is diagnosed of having an increased risk of suffering from memory impairment if the subject's 24, 23, 22, 21, 20, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, nine, eight, seven, six, five, four, three, two or even one of the metabolites herein are at abnormal levels.
[0045] If it is determined that a subject has an increased risk of memory impairment, the attending health care provider may subsequently prescribe or institute a treatment program. In this manner, the present invention also provides for methods of screening individuals as candidates for treatment of memory impairment. The attending healthcare worker may begin treatment, based on the subject's metabolite profile, before there are perceivable, noticeable or measurable signs of memory impairment in the individual.
[0046] Similarly, the invention provides methods of treating memory impairment in a subject in need thereof. The treatment methods include obtaining a subject's metabolite or composite profile as defined herein and prescribing a treatment regimen to the subject if the metabolite and/or composite profile indicate that the subject is at risk of suffering from memory impairment.
[0047] The methods of treatment also include methods of monitoring the effectiveness of a treatment for memory impairment. Once a treatment regimen has been established, with or without the use of the methods of the present invention to assist in a diagnosis of memory impairment, the methods of monitoring a subject's metabolite or composite profile over time can be used to assess the effectiveness of a memory impairment treatment. Specifically, the subject's metabolite or composite profile can be assessed over time, including before, during and after treatments for memory impairment. The metabolite or composite profile can be monitored, with, for example, the normalization or decline in the values of the profile over time being indicative that the treatment may showing efficacy or may lead to improvement in the subjects memory and/or cognitive abilities.
[0048] Suitable treatments for memory impairment that may be initiated in response to an indication that the subject is at risk of suffering from memory impairment include but are not limited to prescribing or administering fenofibtrate, cholinesterase inhibitors and memantine to the subject. Specific examples of cholinesterase inhibitors include but are not limited to donepezil, rivastigmine and galantamine. Another treatment includes but is not limited to SOUVENAID™, which is a mixture of omega-3 fatty acids, choline, uridine monophosphate and a mixture of antioxidants and B vitamins. Other treatments for memory impairment that can be used in conjunction with the present invention include consumption of coconut oil, ginger, green tea, fatty fish, cold-water fish such as but not limited to salmon, tuna, trout, mackerel and sardines, soy products, blueberries and other dark berries, green leafy vegetables and cruciferous vegetables such as but not limited to broccoli.
[0049] All patents and publications mentioned in this specification are indicative of the level of those skilled in the art to which the invention pertains. All patents and publications cited herein are incorporated by reference to the same extent as if each individual publication was specifically and individually indicated as having been incorporated by reference in its entirety.
Examples
[0050] Example 1
[0051] Neurocognitive Methods
[0052] A total of 525 volunteers participated in this study as part of the Rochester/Orange County Aging Study (R/OCAS), an ongoing longitudinal natural history study of cognition in community- dwelling older adults. Briefly, participants were followed with yearly cognitive assessments and blood samples were collected following an overnight fast and withholding of all medications. At baseline and each yearly visit, participants completed clinical assessments, such as activities of daily living, memory complaints, signs and symptoms of depression, and were administered a detailed cognitive assessment.
[0053] For this study, data from the cognitive tests were used to classify participants into groups for biomarker discovery. Standardized scores (Z-scores) were derived for each participant on each cognitive test and the composite Z-scores were computed for five cognitive domains (attention, executive, language, memory, visuoperceptual) (Table 2).
Table 2
[0054] Normative data for Z-score calculations were derived from the performance of the participants on each of the cognitive tests adjusted for age, education, sex, and visit. To reduce the effect of cognitively impaired participants on the mean and SD, age-, education-, sex, and visit- adjusted residuals from each domain Z-score model were robustly standardized to have median 0
and robust SD=1, where the robust SD=IQ /1.35, as 1.35 is the IQR (Inter-Quartile Range) of a standard normal distribution.
[0055] The participants were then categorized into groups of incident aMCI or early AD (combined into one category aMCI/AD), cognitively normal control (NC), and those who converted to MCI or AD over the course of the study (Converters) based on these composite scores. Impairment was defined as a Z-score 1.35 SD below the cohort median. All participants classified as aMCI met recently revised criteria for the amnestic subtype of MCI. Other behavioral phenotypes of MCI were excluded to concentrate on the amnestic form, which most likely represents nascent Alzheimer's pathology. All early AD participants met recently revised criteria for probable Alzheimer's disease with impairment in memory and at least one other cognitive domain. For the MCI and early AD groups, scores on the measures of memory complaints (MMQ) and activities of daily living (PGC- IADL) were used to corroborate research definitions of these states. All Converters had non- impaired memory at entry to the study (Zmem >-1.35), developed memory impairment over the course of the study (Zmem <-l-35) and met criteria for the above definitions of aMCI or AD. To enhance the specificity of the biomarker analyses, NC participants in this study were conservatively defined with Zmem +1 SD of the cohort median rather than simply > -1.35, and all other Z-scores > - 1.35 SD.
[0056] For each subject, Zmem(last), Zatt(last), Zexe(last), Z|an(last), and Zvis(last) were defined as the age-gender-education-visit-adjusted robust Z-scores for the last available visit for each subject. The aMCI/AD group was defined as those participants whose adjusted Zmem was 1 IQR below the median at their last available visit, i.e., Zmem(last) < -1.35. Converters were defined as that subset of the a MCI/AD group whose adjusted Zmem at baseline visit 0 was no more than 1 IQR below the median, i.e., Zmem(visit=0) > -1.35 and Zmem(last) < -1.35. Participants were classified as NC if they had central scores on all domains at both the first and last visits, i.e., only if they met all of the following six conditions: (i) -1 < Zmem(last) < 1, (ii) -1 < Zmem(visit=0) < 1, (iii) Zmin(last) > -1.35, (iv) Zmin(visit=0) > -1.35, (v) Zmax(last) < 1.35, and (vi) Zmax(visit=0) < 1.35, where Zmax(last) and Zmax(visit=0) denote the maximum of the five adjusted Z-scores at the last and first visits, respectively. Zmem for normal participants had to be within 0.74 IQR (1 SD) of the median, rather than just 1 IQR (1.35 SD), to guarantee that they were > 0.25 IQR (0.35 SD) from aMCI/AD participants.
[0057] After three years of being in the study, (December, 2010), 202 participants had completed a baseline and two yearly visits. At the third visit, 53 participants met criteria for aMCI/AD and 96 met
criteria for NC. Of the 53 aMCI/AD participants, 18 were Converters and 35 were incident aMCI or AD. The remaining participants did not meet the criteria for either group and were not considered for biomarker profiling. Some of these individuals met criteria for non-amnestic MCI and many had borderline or even above average memory scores that precluded their inclusion as either aMCI/AD or NC. 53 of the NC participants were matched to the 53 aMCI/AD participants based on sex, age, and education level. Blood samples were obtained on the last available study visit for the 53 MCI/AD and the 53 NC for biomarker discovery. Two blood samples from each of the 18 Converters were also included: one from the baseline visit (Cpre) when Zmem was non-impaired and one from the third visit (Cpost) when Zmem was impaired and they met criteria for either aMCI or AD. Thus, at total of 124 samples from 106 participants were analyzed.
[0058] Internal cross-validation was employed to validate findings from the discovery phase. Blood samples for validation were identified at the end of the fifth year of the study and all 106 participants included in the discovery phase were excluded from consideration for the validation phase. Cognitive composite Z-scores were re-calculated based on the entire sample available and the same procedure and criteria were used to identify samples for the validation phase. A total of 145 participants met criteria for a group: 21aMCI/AD and 124 NC. Of the 21 aMCI/AD, 10 were Converters. 20 of the NC participants were matched to the aMCI/AD participants on the basis of age, sex, and education level as in the discovery phase. In total, 41 participants contributed samples to the validation phase and, as before, the 10 Converters also contributed a baseline sample (Cpre) for a total of 51 samples.
[0059] Neurocognitive Statistical Analyses
[0060] The neurocognitive analyses were designed to demonstrate the general equivalence of the discovery and validation samples on clinical and cognitive measures. Separate Multivariate Analysis of Variance (MANOVA's) tests were used to examine discovery/validation group performance on the composite Z-scores and on self-report measures of memory complaints, memory related functional impairment, depressive symptoms, and a global measure of cognitive function. In the first
MANOVA, biomarker sample (discovery, validation) was the independent variable and M MQ, IADL, GDS, and MMSE were the dependent variables. In the second MANOVA, biomarker sample
(discovery, validation) was the independent variable and the five cognitive domain Z-scores (Zatt, Zexe, Zian, Zmem, and Zvis) were the dependent variables. Significance was set at alpha=0.05 and Tukey's HSD procedure was used for post-hoc comparisons. All statistical analyses were performed using SPSS (version 21).
[0061] Example 2
[0062] Reagents
[0063] LC/MS-grade acetonitrile (ACN), Isopropanol (IPA), water and methanol were purchased from Fisher Scientific (New Jersey, USA). High purity formic acid (99%) was purchased from Thermo- Scientific (Rockford, IL). Debrisoquine, 4-Nitrobenzoic acid (4-NBA), Pro-Asn, Glycoursodeoxycholic acid, Malic acid, were purchased from Sigma (St. Louis, MO, USA). All lipid standards including 14:0 LPA, 17:0 Ceramide, 12:0 LPC, 18:0 Lyso PI and PC(22:6/0:0) were procured from Avanti Polar Lipids Inc. (USA).
[0064] Metabolite extraction
[0065] Briefly, the plasma samples were thawed on ice and vortexed. For metabolite extraction, 25 μί of plasma sample was mixed with 175 μί of extraction buffer (25% acetonitrile in 40% methanol and 35% water) containing internal standards [10 μί of debrisoquine (1 mg/mL), 50 μί of 4, nitro- benzoic acid (1 mg/mL), 27.3 μΙ of Ceramide (1 mg/mL) and 2.5 μί of LPA (lysophosphatidic acid) (4 mg/mL) in 10 mL). The samples were incubated on ice for 10 minutes and centrifuged at 14,000 rpm at 4°C for 20 minutes. The supernatant was transferred to a fresh tube and dried under vacuum. The dried samples were reconstituted in 200 μί of buffer containing 5% methanol, 1% acetonitrile and 94% water. The samples were centrifuged at 13,000 rpm for 20 minutes at 4°C to remove fine particulates. The supernatant was transferred to a glass vial for UPLC-ESI-Q-TOF-MS analysis.
[0066] Each sample (2 μί) was injected onto a reverse-phase CSH C18 1.7 μΜ 2.1x100 mm column using an Acquity H-class UPLC system (Waters Corporation, USA). The gradient mobile phase comprised of water containing 0.1% formic acid solution (Solvent A), 100% acetonitrile (Solvent B) and 10% acetonitrile in isopropanol (IPA) containing 0.1% formic acid and lOmM ammonium formate (Solvent C). Each sample was resolved for 13 minutes at a flow rate of 0.5 mL/min for 8 min and then 0.4 mL/min from 8 to 13 min. The UPLC gradient consisted of 98% A and 2% B for 0.5 min then a ramp of curve 6 to 60% B and 40% A from 0.5 min to 4.0 min, followed by a ramp of curve 6 to 98% B and 2% A from 4.0 to 8.0 min, then ramped to 5% B and 95% C from 9.0 min to 10.0 min at a flow rate of 0.4 ml/min, and finally to 98% A and 2% B from 11.0 min to 13 minutes. The column eluent was introduced directly into the mass spectrometer by electrospray ionization. Mass spectrometry was performed on a Quadrupole-Time of Flight (Q-TOF) instrument (Xevo G2 QTOF, Waters Corporation, USA) operating in either negative (ESI ) or positive (EST) electrospray ionization mode with a capillary voltage of 3200 V in positive mode and 2800 V in negative mode, and a sampling cone voltage of 30 V in both modes. The desolvation gas flow was set to 750 L h"1 and the
temperature was set to 350°C while the source temperature was set at 120°C. Accurate mass was maintained by introduction of a lock spray interface of leucine-enkephalin (556.2771 [M+H]+ or 554.2615 [M-H]") at a concentration of 2 pg/μΙ in 50% aqueous acetonitrile and a rate of 2 μΙ/min. Data were acquired in centroid MS mode from 50 to 1200 m/z mass range for TOF-MS scanning as single injection per sample and the batch acquisition was repeated to check experimental reproducibility. For the metabolomics profiling experiments, pooled quality control (QC) samples (generated by taking an equal aliquot of all the samples included in the experiment) were run at the beginning of the sample queue for column conditioning and every ten injections thereafter to assess inconsistencies that are particularly evident in large batch acquisitions in terms of retention time drifts and variation in ion intensity over time. This approach has been recommended and used as a standard practice by leading metabolomics researchers. A test mix of standard metabolites was run at the beginning and at the end of the run to evaluate instrument performance with respect to sensitivity and mass accuracy. The sample queue was randomized to remove bias.
[0067] Stable Isotope Dilution - Multiple Reaction Monitoring Mass Spectrometry (SID-MRM-MS)
[0068] Targeted metabolomic analysis of plasma sample was performed using the Biocrates Absolute-IDQ P180 (BIOCRATES, Life Science AG, Innsbruck, Austria). This validated targeted assay allows for simultaneous detection and quantification of metabolites in plasma samples (10 μί) in a high throughput manner. The methods have been described in detail. The plasma samples were processed as per the instructions by the manufacturer and analyzed on a triple quadrupole mass spectrometer (Xevo TQ.-S, Waters Corporation, USA) operating in the MRM mode. The
measurements were made in a 96 well format for a total of 148 samples, seven calibration standards and three quality control samples were integrated in the kit.
[0069] Briefly, the flow injection analysis (FIA) tandem mass spectrometry (MS/MS) method was used to quantify a panel of 144 lipids simultaneously by multiple reaction monitoring. The other metabolites were resolved on the UPLC and quantified using scheduled MRMs. The kit facilitated absolute quantitation of 21 amino acids, hexose, carnitine, 39 acylcarnitines, 15 sphingomyelins, 90 phosphatidylcholines and 19 biogenic amines. Data pre-processing was performed using the MetlQ software (Biocrates) while the statistical analyses were performed using linear regression as well as the STAT pack module v3 (Biocrates). The concentration is expressed as nmol/L. Quality control samples were used to assess reproducibility of the assay. The mean of the coefficient of variation (CV) for the 180 metabolites was 0.08 and 95 % of the metabolites had a CV of <0.15.
[0070] Metabolite Statistical Analyses
[0071] The m/z features of metabolites were normalized with log transformation that stabilized the variance followed with a quantile normalization to make the empirical distribution of intensities the same across samples. The metabolites were selected among all those known to be identifiable using a ROC regularized learning technique, based on the least absolute shrinkage and selection operator (LASSO) penalty as implemented with the R package 'gimnet', which uses cyclical coordinate descent in a pathwise fashion. The regularization path over a grid of values was obtained for the tuning parameter lambda through 10-fold cross-validation. The optimal value of the tuning parameter lambda, which was obtained by the cross-validation procedure, was then used to fit the model. Selection criteria for the 24 metabolite panel included: 1) the concordantly expressed (discovery and validation) biomarkers from the 10 lipid panel (propionyl AC, lyso PC a C18:2, PC aa C36:6, C16:l-OH (Hydroxyhexadecenoyl-L-carnitine), PC aa C38:0, PC aa 36:6, PC aa C40:l, PC aa C40:2, PC aa C40:6 and PC ae C40:6), which were previously described in International PCT Application Publication No. WO 2015/061317 and Mapstone, M., et al., Nat. Med., 20:415-418 (2014), both of which are incorporated by reference; 2) any annotated significant (p<0.05) metabolite species; 3) newly identified metabolites that were concordantly expressed (either increased or decreased). With these criteria, the smallest biomarker panel that provided an ROC area under the curve (AUC) >0.99 was the 24-member panel identified herein.
[0072] Results
[0073] Over the course of the study, 74 participants met criteria for either aMCI or mild AD, 46 of these were incidental cases at entry and 28 phenoconverted (Converters) from a non-impaired memory status at entry (Cpre). The average time to phenoconversion was 2.1 years (range = 1-5 years). 53 aMCI/AD participants were selected, including 18 Converters, and 53 age-, education-, and sex-matched cognitively normal control (NC) participants for untargeted lipidomics biomarker discovery. Internal cross validation was used to evaluate the accuracy of the discovered metabolite profile in classifying a blinded sample of 51 subjects consisting of the remaining subset of 21 aMCI/AD participants, including 10 Converters, and an additional 20 matched NC.
[0074] The aMCI/AD, Converter, and NC groups were defined primarily using a composite measure of memory performance in addition to composite measures of other cognitive abilities and clinical measures of memory complaints and functional capacities. (See Tables 2 and 3).
Table 3
[0075] To confirm the reproducibility of the metabolite panel from the discovery samples, targeted quantitative metabolomics analyses was performed using plasma from the independent validation group of 40 participants as an independent cross validation. One sample from the aMCI/AD group was not available for metabolomic analysis. The validation samples were obtained from the last available visit from the aMCI/AD, Cpre, and NC groups and were designated as groups A, B, and C and analyzed in a blinded fashion, without specification of diagnostic identities. The samples were processed and analyzed using the same SID-MRM-MS technique as in the discovery phase. The blinded data were statistically analyzed to determine if the unknown groups could be characterized into the correct diagnostic categories based solely on the levels of the metabolite panel.
[0076] The metabolomic data was used from the untargeted LASSO analysis to build separate classifier models that would distinguish the NC group from the Cpre group. A receiver operating characteristic (ROC) analysis was employed to determine the a rea under the ROC curve (AUC) to assess the performance of the classifier models in differentiating the groups. When distinguishing between the Cpre and NC groups, the LASSO identified metabolites yielded an AUC of 0.995. This high AUC value demonstrates robust discrimination between the Cpre and NC groups by the models.
[0077] Example 3
[0078] Methods for Preclinical Alzheimer's Disease Index (PADI)
[0079] All participants were community-dwelling, older adults from the greater Rochester, NY and Irvine, CA communities. Participants were recruited through local media (newspaper and television advertisements), senior organizations, and word of mouth. Inclusion criteria included age 70 or older, proficiency with written and spoken English and corrected vision and hearing necessary to complete the cognitive battery. Participants were excluded for the presence of known major psychiatric or neurological illness (including Alzheimer's disease or MCI, cortical stroke, epilepsy, and psychosis) at time of enrollment, current or recent (< 1 month) use of anticonvulsants, neuroleptics, HAART, antiemetics, and antipsychotics for any reason, and serious blood diseases including chronic abnormalities in complete blood count and anemia requiring therapy and/or transfusion.
[0080] Operationalizing groups
[0081] Alzheimer's disease is a cognitive disorder defined in numerous diagnostic schemes by a primary impairment in memory. The classification of preclinical Alzheimer's disease via PADI relies
heavily on the assessment of memory, but it is recognized that the preclinical state may also involve other cognitive domains, thus the methods disclosed herein rely on assessing all primary cognitive domains (Table 4). The primary Memory outcome was based on the Rey Auditory Visual Learning Test (RAVLT). Learning was defined as the sum of the number of correct words recalled over the 5 learning trials; Retrieval was defined as the total number of correct words recalled from trial A7; and Retention was defined as the total number correct words recognized minus the number of false positives. Each of these three sub-scores (Learning, Retrieval, and Retention) was converted to an empirical Z-score by subtracting its sample mean and dividing by its sample SD. The three resulting positively correlated RAVLT-based Z-scores were then averaged to form the composite Memory score (Zmem). So defined, Zmem was approximately normal with mean 0, SD<1, and some negative skewness attributable to the fact that healthy participants often score at or near the upper bounds, especially for Retention. The Attention (Zatt), Executive (Zexe), and Language (Zlan) scores were each defined as averages of empirical Z-score transforms of pairs of subscores, as follows. Zatt: (1) completion time (in seconds, truncated at 300s) for the Trail Making Test Part A and (2) the
Wechsler Memory Scale-Ill Forward Digit Span; Zexe: (1) completion time (in seconds, truncated at 300s) for the Trail Making Test Part B and (2) the Wechsler Memory Scale-Ill Backward Digit Span; Zlan: (1) total score (out of 60) for the Boston Naming Test and (2) Category Fluency (Animals Named in 1 minute). The Visuoperceptual score (Zvis) was simply the empirical Z-transform of the Total Score (out of 30) for the Hooper Visual Organization Test (HVOT), which itself was approximately normally distributed. See Table 2 for composite Z score components.
[0082] Standardization and Adjustment for Age, Gender, Education, and Visit
[0083] Linear Generalized Estimating Equation (GEE) models were used to model each of the five cognitive domains (Memory, Attention, Executive, Language, and Visuospatial) as a function of age, gender, education, and visit number, using a homoscedastic working independence covariance matrix, i.e., least squares was used to estimate all linear model coefficients based on the pooled data from all available visits. Since Memory was the focal cognitive domain, the functional form for the covariates was selected to best model Memory, and exactly the same functional form was used for the other four cognitive domains. Given the evidence of nonlinearity, the effect of education was modeled using a continuous piecewise linear spline with knots at 12, 14, and 16 years of education. Memory increased with years of education between 5-12 years, decreased between 12-14 years, increased again between 14-16 years, and then nearly leveled off for 16-23 years. Age was modeled linearly, as there was insufficient evidence to support nonlinearity via piecewise linear spline knots or a quadratic component, yet Memory scores decreased linearly with age. Visit was modeled via an
indicator for baseline visit 0, given that there was insufficient evidence that subsequent visits differed from each other, yet there was evidence that subsequent visits had higher Memory scores compared with baseline visit 0. Gender was modeled via an indicator for males, who had lower memory scores than females. There was insufficient evidence to support any interactions. Residuals from each model were then robustly standardized to have median 0 and robust SD=1, where the robust SD = IQR/1.35, as 1.35 is the IQR (Inter-Quartile Range) of a standard normal distribution. The choice to use robust measures of location (median) and dispersion (IQR) was made to reduce the influence that cognitively impaired participants might have on the mean and SD, and in recognition of the slight negative skewness of the residuals. The robustly standardized residuals were then viewed as age-gender-education-visit-adjusted robust Z-scores for each of the five cognitive domains.
[0084] Defining aMCI/AD, Converters, and NC
[0085] For each subject, Zmem(last), Zatt(last), Zexe(last), Zlan(last), and Zvis(last) were defined as the age-gender-education-visit-adjusted robust Z-scores for the last available visit for each subject. The aMCI/AD group was defined as those participants whose adjusted Zmem was 1 IQR below the median at their last available visit, i.e., Zmem(last) < -1.35. Converters were defined as that subset of the MCI/AD group whose adjusted Zmem at baseline visit 0 was no more than 1 IQR below the median, i.e., Zmem(visit=0) > -1.35 and Zmem(last) < -1.35. Participants were classified as NC if they had central scores on all domains at both the first and last visits, i.e., only if they met all of the following six conditions: (i) -1 < Zmem(last) < 1, (ii) -1 < Zmem(visit=0) < 1, (iii) Zmin(last) > -1.35, (iv) Zmin(visit=0) > -1.35, (v) Zmax(last) < 1.35, and (vi) Zmax(visit=0) < 1.35, where Zmax(last) and Zmax(visit=0) denote the maximum of the five adjusted Z-scores at the last and first visits, respectively. Zmem for normal participants had to be within 0.74 IQR (1 SD) of the median, rather than just 1 IQR (1.35 SD), to guarantee that they were > 0.25 IQR (0.35 SD) from aMCI/AD participants.
[0086] Frequency matching aMCI/AD and NC on age, education, and sex
[0087] Comparing the distributions of age, education, and sex for participants classified as aMCI/AD and NC it was observed that NC participants were younger. Given this, and given budget limitations for genetic profiling, NC participants were frequency matched to the aMCI/AD participants for each of the Discovery and Validation samples based on age, education, and sex. This was accomplished by stratifying the sample by four groups of age at the last visit (75-79, 80-85, 86-94, 95-100), three education groups (5-12, 13-18, 19-23), and two sexes (male, female), resulting in 24 age-education-
sex strata. An equal number of NC participants were randomly selected to match the aMCI/AD participants per the strata for the Discovery sample, and because one of the strata did not contain a sufficient number of NC participants, the Validation sample had one less NC subject than the aMCI/AD group.
[0088] Derivation of the Preclinical Alzheimer's Disease Index (PADI)
[0089] For the development of the preclinical Alzheimer's disease Index (PADI), only cognitively normal participants were included at entry (+0.7 SD from the group mean on all domain z-scores) who subsequently either phenoconverted to amnestic MCI or LOAD (Converters, n = 28) or remained cognitively normal (NC, n = 73) over the course of the study. Time to phenoconversion averaged 2.1 years. The baseline characteristics of the Converter and NC groups were similar except in three cognitive domains: attention, executive function, and memory (Table 4). Stepwise logistic regression revealed that only the attention (OR = 0.247, 95% CI: 0.09 - 0.68) and memory (OR = 0.232, 95% CI: 0.10 - 0.54) z-scores were significant predictors of group. The PADI was constructed by combining the two domains (weighted -0.484 for attention and -0.516 for memory) such that higher scores indicate poorer cognitive performance.
[0090] The utility of the PADI was tested in the remaining study participants (n=515), to identify phenoconversion risk in subjects not impaired at baseline, but who exhibited subtle deficits that did not allow their consideration as "normal". A total of 35 of the latter met criteria for MCI or AD an average of 1.9 years later. The baseline PADI distinguished these at risk individuals from the rest (ROC AUC = 0.93, 95% CI: 0.90, 0.97). Survival analysis, controlling for age, years of education, APOE4 status, global cognition, depression, and subjective memory complaints was also significant (HR = 4.57, 95% CI = 3.47, 6.01) (Figure 4).
Attorney Docket No. 036681-
TABLE 4 Cognitive Measures: Means and Standard Deviations
Attorney Docket No. 036681-
TABLE 5 PADI Baseline Demographic and Health Characteristics.
Claims
1. A method of determining if a subject has an increased risk of suffering from memory
impairment, the method comprising a) analyzing at least one sample from the subject to determine a value of the subject's metabolite profile, and b) comparing the value of the subject's metabolite profile with the value obtained from subjects determined to define a normal metabolite profile, to determine if the subject's metabolite profile is altered compared to a normal metabolite profile, wherein a change in the value of the subject's metabolite profile is indicative that the subject has an increased risk of suffering from future memory impairment compared to those defined as having a normal metabolite profile.
2. The method of claim 1, wherein the metabolite profile comprises one or more components selected from the group consisting of C3, lysoPC a C18:2, PC aa C36:6, PC aa C38:0, PC aa C38:6, PC aa C40:l, PC ae C40:6, C12:l, PC aa C32:0, C16:2, PC aa C40:5, C5-OH/C3-DC-M, PC ae C34:0, C1S:1-0H, PC aa C34:4, Asparagine (Asn), Asymmetric dimethylarginine (ADMA), C5, PC aa C38:3, C10:l, PC ae C36:4, C10:2, C9 and PC ae C42:l.
3. The method of claims 1 or 2, wherein the normal metabolite profile comprises the subject's metabolite profile prior to the onset of memory impairment.
4. The method of any of claim 1-3, wherein the normal metabolite profile comprises a
metabolite profile generated from a population of individuals that do not presently or in the future display memory impairment.
5. A method of monitoring the progression of memory impairment in a subject, the method comprising a) analyzing at least two blood samples from the subject with each sample taken at
different time points to determine the values of each of the subject's metabolite profile, and b) comparing the values of the subject's metabolite profile over time to determine if the subject's metabolite profile is changing over time,
wherein a change in the subject's metabolite profile over time is indicative that the subject's risk of suffering from memory impairment is increasing over time.
6. The method of claim 5, wherein the metabolite profile comprises one or more components selected from the group consisting of C3, lysoPC a C18:2, PC aa C36:6, PC aa C38:Q, PC aa C38:6, PC aa C40:l, PC ae C4Q:6, C12:l, PC aa C32:0, C16:2, PC aa C40:5, C5-OH/C3-DC-M, PC ae C34:0, C18:l-OH, PC aa C34:4, Asparagine (Asn), Asymmetric dimethylarginine (ADMA), C5, PC aa C38:3, C10:l, PC ae C36:4, C10:2, C9 and PC ae C42:l.
7. A method of monitoring the progression of a treatment for memory impairment in a subject, the method comprising a) analyzing at least two samples from a subject undergoing treatment for memory
impairment with each sample taken at different time points to determine the values of each of the subject's metabolite profile, and b) comparing the values of the subject's metabolite profile over time to determine if the subject's metabolite profile is changing over time in response to the treatment, wherein a lack of change or a further deviation from a normal metabolite profile in the subject's metabolite profile is indicative that the treatment for memory impairment is not effective, and wherein an approximation of the subject's metabolite profile over time towards a normal metabolite profile is indicative that the treatment for memory impairment is effective in treating memory impairment in the subject.
8. The method of claim 7, wherein the metabolite profile comprises one or more components selected from the group consisting of C3, lysoPC a C18:2, PC aa C36:6, PC aa C38:0, PC aa C38:6, PC aa C40:l, PC ae C40i6, ClZil, PC aa C32:0, C16:2, PC aa C40:5, C5-OH/C3-DC- , PC ae C34:0, C18:l-OH, PC aa C34:4, Asparagine (Asn), Asymmetric dimethylarginine (ADMA), C5, PC aa C38:3, C10:l, PC ae C36:4, C10:2, C9 and PC ae C42:l.
9. A method of determining if a subject has an increased risk of suffering from memory
impairment, the method comprising analyzing at least one sample from the subject to determine levels of individual metabolites and comparing the levels of individual metabolites with the value of levels of the metabolites in one or more normal individuals to determine if the levels of each metabolite are altered compared to normal levels, wherein a
change in the value of the subject's metabolites is indicative that the subject has an increased risk of suffering from memory impairment compared to a normal individual.
10. The method of claim 9 wherein the metabolites are selected from the group consisting of
C3, lysoPC a C18:2, PC aa C36:6, PC aa C38:0, PC aa C38:6, PC aa C40:l, PC ae C4Q:6, C12:l, PC aa C32:0, C16:2, PC aa C40:5.. C5-OH/C3-DC-M, PC ae C34:0, C18:l-OH, PC aa C34:4, Asparagine (Asn), Asymmetric dimethyiarginine (ADM A), C5, PC aa C38:3, C1Q:1, PC ae C36:4, C10:2, C9 and PC ae C42:l.
11. A method of measuring levels of metabolites in a subject, the method comprising obtaining a sample from a subject at a first time point and determining individual levels of C3, lysoPC a C18:2, PC aa C36:6, PC aa C38:0, PC aa C38i6, PC aa C40:l, PC ae C40:6, C12:l, PC aa C32:0, C16:2; PC aa C40:5, C5-OH/C3-DC-M, PC ae C34:0, C18:l-OH, PC aa C34:4, Asparagine (Asn), Asymmetric dimethyiarginine (AD A), C5, PC aa C38:3, C1Q:1, PC ae C36:4, C10:2, C9 and PC ae C42il at said first time point.
12. A method of determining if a subject has an increased risk of suffering from memory
impairment, the method comprising a) generating a plasma metabolite index (PMI) for the subject comprising determining individual plasma levels of the metabolites C3. lysoPC a C18:2, PC aa C36:6, PC aa C38:0, PC aa C38:6, PC aa C40:l, PC ae C40:6, C12il, PC aa C32:0, C16:2, PC aa C40:5; C5- OH/C3-DC-M, PC ae C34:0, C18:l-OH, PC aa C34:4, Asparagine (Asn), Asymmetric dimethyiarginine (ADMA), C5; PC aa C38:3, C10:l, PC ae C36:4, C10:2, C9 and PC ae C42il in the subject at a first time point, and b) comparing the generated PMI to a normal PMI for the same metabolites to determine if the subject's metabolite profile is altered compared to the normal PMI, c) wherein a change in the subject's PMI for the metabolites compared to the normal PMI is indicative that the subject has an increased risk of suffering from future memory impairment compared to subjects that have a normal PMI.
13. The method of claim 12, wherein a decrease in the value of the subject's PMI for the
metabolites compared to the normal PMI is indicative that the subject has an increased risk of suffering from future memory impairment compared to subjects that have a normal PMI.
A method of determining if a subject has an increased risk of suffering from memory impairment, the method comprising a) analyzing at least one sample from the subject to determine a value of the subject's metabolite profile, and b) comparing the value of the subject's metabolite profile with the value obtained from subjects determined to define a normal metabolite profile, to determine if the subject's metabolite profile is altered compared to a normal metabolite profile, wherein the subject's Pre-Clinical Alzheimer's Disease Index (PADI) indicates that the subject is at risk of suffering from future memory impairment prior to determining the subject's metabolite profile, and wherein a change in the value of the subject's metabolite profile is indicative that the subject has an increased risk of suffering from future memory impairment compared to those defined as having a normal metabolite profile.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US15/575,115 US10900980B2 (en) | 2015-05-18 | 2016-05-18 | Metabolic biomarkers for memory loss |
US17/157,932 US11808774B2 (en) | 2015-05-18 | 2021-01-25 | Metabolic biomarkers for memory loss |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US201562163034P | 2015-05-18 | 2015-05-18 | |
US62/163,034 | 2015-05-18 | ||
US201562194223P | 2015-07-18 | 2015-07-18 | |
US62/194,223 | 2015-07-18 |
Related Child Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/575,115 A-371-Of-International US10900980B2 (en) | 2015-05-18 | 2016-05-18 | Metabolic biomarkers for memory loss |
US17/157,932 Continuation US11808774B2 (en) | 2015-05-18 | 2021-01-25 | Metabolic biomarkers for memory loss |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016187317A1 true WO2016187317A1 (en) | 2016-11-24 |
Family
ID=57320684
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2016/033100 WO2016187317A1 (en) | 2015-05-18 | 2016-05-18 | Metabolic biomarkers for memory loss |
Country Status (2)
Country | Link |
---|---|
US (2) | US10900980B2 (en) |
WO (1) | WO2016187317A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018144816A1 (en) * | 2017-02-02 | 2018-08-09 | Georgetown University | Metabolic biomarkers for cognitive ability |
US11808774B2 (en) | 2015-05-18 | 2023-11-07 | Georgetown University | Metabolic biomarkers for memory loss |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20220178954A1 (en) * | 2019-03-08 | 2022-06-09 | Duke University | Stratification by sex and apoe genotype identifies metabolic heterogeneity in alzheimer's disease |
CN113484453B (en) * | 2021-07-07 | 2022-08-09 | 天津中医药大学 | Cerebral arterial thrombosis early warning method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7314444B2 (en) * | 2002-01-25 | 2008-01-01 | Albert Einstein College Of Medicine Of Yeshiva University | Memory assessment by retrieval speed and uses thereof |
US20120015357A1 (en) * | 2008-11-27 | 2012-01-19 | Karsten Suhre | Prediction of lipid-metabotype-related physiological susceptibilities |
US20120214870A1 (en) * | 2009-08-31 | 2012-08-23 | University Of South Australia | Methods for diagnosis and treatment of chronic fatigue syndrome |
WO2015061317A1 (en) * | 2013-10-21 | 2015-04-30 | Georgetown University | Biomarkers for memory loss |
Family Cites Families (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1998040748A1 (en) | 1997-03-14 | 1998-09-17 | Neuromark | Diagnosing neurologic disorders |
US8637321B2 (en) | 2005-10-24 | 2014-01-28 | Duke University | Lipidomics approaches for central nervous system disorders |
EP2052255A2 (en) | 2006-08-18 | 2009-04-29 | Huntington Medical Research Institutes | Methods of determining levels of free amino acids and dipeptides and diagnosing alzheimer's diseases |
US8026099B2 (en) | 2007-07-26 | 2011-09-27 | Washington University | Lipid profile as a biomarker for early detection of neurological disorders |
WO2011005893A2 (en) | 2009-07-07 | 2011-01-13 | Abbott Laboratories | Biomarkers and methods for detecting alzheimer's disease |
ES2581178T3 (en) | 2009-07-29 | 2016-09-01 | Pharnext | New diagnostic tools for Alzheimer's disease |
GB201021502D0 (en) | 2010-12-20 | 2011-02-02 | Cambridge Entpr Ltd | Biomarkers |
US9529002B2 (en) | 2011-05-18 | 2016-12-27 | Balwant Rai | Biomarkers for prediction, diagnosis, and monitoring of Alzheimer's disease |
FI20115576A0 (en) | 2011-06-10 | 2011-06-10 | Teknologian Tutkimuskeskus Vtt Oy | A method for diagnosing Alzheimer's disease |
WO2013119651A1 (en) | 2012-02-08 | 2013-08-15 | Nestec Sa | Methods for diagnosing cognitive impairment |
WO2015183917A2 (en) | 2014-05-27 | 2015-12-03 | Georgetown University | Metabolic biomarkers for memory loss |
US10900980B2 (en) * | 2015-05-18 | 2021-01-26 | Georgetown University | Metabolic biomarkers for memory loss |
US10900977B2 (en) | 2015-10-16 | 2021-01-26 | Georgetown University | Protein biomarkers for memory loss |
US11280797B2 (en) | 2017-02-02 | 2022-03-22 | Georgetown University | Metabolic biomarkers for cognitive ability |
-
2016
- 2016-05-18 US US15/575,115 patent/US10900980B2/en active Active
- 2016-05-18 WO PCT/US2016/033100 patent/WO2016187317A1/en active Application Filing
-
2021
- 2021-01-25 US US17/157,932 patent/US11808774B2/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7314444B2 (en) * | 2002-01-25 | 2008-01-01 | Albert Einstein College Of Medicine Of Yeshiva University | Memory assessment by retrieval speed and uses thereof |
US20120015357A1 (en) * | 2008-11-27 | 2012-01-19 | Karsten Suhre | Prediction of lipid-metabotype-related physiological susceptibilities |
US20120214870A1 (en) * | 2009-08-31 | 2012-08-23 | University Of South Australia | Methods for diagnosis and treatment of chronic fatigue syndrome |
WO2015061317A1 (en) * | 2013-10-21 | 2015-04-30 | Georgetown University | Biomarkers for memory loss |
Non-Patent Citations (2)
Title |
---|
KIM ET AL.: "Aging-related changes in mouse serum glycerophospholipid profiles", OSONG PUBLIC HEALTH RES PERSPECT, vol. 5, no. 6, December 2014 (2014-12-01), pages 345 - 350, XP055330989 * |
RUZZO ET AL.: "Deficiency of asparagine synthetase causes congenital microcephaly and a progressive form of encephalopathy", NEURON, vol. 80, no. 2, 16 October 2013 (2013-10-16), pages 429 - 441, XP028757374 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11808774B2 (en) | 2015-05-18 | 2023-11-07 | Georgetown University | Metabolic biomarkers for memory loss |
WO2018144816A1 (en) * | 2017-02-02 | 2018-08-09 | Georgetown University | Metabolic biomarkers for cognitive ability |
US11280797B2 (en) | 2017-02-02 | 2022-03-22 | Georgetown University | Metabolic biomarkers for cognitive ability |
Also Published As
Publication number | Publication date |
---|---|
US20210285973A1 (en) | 2021-09-16 |
US20180149665A1 (en) | 2018-05-31 |
US10900980B2 (en) | 2021-01-26 |
US11808774B2 (en) | 2023-11-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11808774B2 (en) | Metabolic biomarkers for memory loss | |
JP6021187B2 (en) | Metabolic biomarkers of autism | |
EP3019624B1 (en) | Biomarkers of autism spectrum disorder | |
US20210405074A1 (en) | Biomarkers for the diagnosis and characterization of alzheimer's disease | |
ES2927921T3 (en) | Methods using metabolite panels for the diagnosis and differentiation of neonatal encephalopathy | |
US20210325409A1 (en) | Biomarkers and uses thereof for diagnosing the silent phase of alzheimer's disease | |
US20150090010A1 (en) | Method for diagnosing heart failure | |
US20170052204A1 (en) | Metabolic and Genetic Biomarkers for Memory Loss | |
Schulte et al. | Salivary metabolite levels in perinatally HIV-infected youth with periodontal disease | |
US10578629B2 (en) | Biomarkers for memory loss | |
US20200363436A1 (en) | Methods of assessing future memory loss by measuring polyunsaturated fatty acids | |
US10890589B2 (en) | Metabolic biomarkers for memory loss | |
US20200081019A1 (en) | Identification of signatures for neurodegeneration diseases diagnoses | |
US20170184613A1 (en) | Exosome and lipid biomarkers for memory loss | |
US11280797B2 (en) | Metabolic biomarkers for cognitive ability | |
KR20230163420A (en) | Method for evaluating the risk of future onset of dementia | |
WO2017066739A1 (en) | Protein biomarkers for memory loss | |
Mares et al. | Brain and serum lipidomic profiles implicate Lands cycle acyl chain remodeling association with APOEε4 and mild cognitive impairment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16797224 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 16797224 Country of ref document: EP Kind code of ref document: A1 |